SYSTEMS AND METHODS FOR QUANTIFICATION AND CLASSIFICATION OF FLUIDS IN HUMAN CAVITIES IN ULTRASOUND IMAGES

Abstract

Ultrasound imaging systems and methods are disclosed. In one embodiment, an ultrasonography method includes creating a database that is representative of a tissue, a fluid, or a cavity of a body, and transmitting ultrasound pulses into a region-of-interest in a patient. Echoes are received from the region of interest, and based upon the received echoes, compiling an ultrasonic pattern of the region-of-interest is compiled. The pattern is processed by comparing the region-of-interest patterns to the pattern information stored in the database. A composition within the region-of-interest of the patient is then determined.

Description

FIELD OF INVENTION

This invention relates to ultrasound imaging of bodily tissues, bodily fluids, and fluid-filled cavities.

BACKGROUND OF THE INVENTION

The following applications are incorporated by reference as if fully set forth herein: U.S. application Ser. Nos. 11/213,284 filed on Aug. 26, 2005; 11/010,539 filed on Dec. 13, 2004; and 10/523,681 filed on Feb. 3, 2005.

Ultrasound imaging is accomplished by placing an ultrasound transducer on a selected location of a body and projecting ultrasound energy into the body. Acoustic waves reflecting from internal structures in the body are then received by the transducer and are processed to form an image of the internal structures. In a particular ultrasound method, amplitudes of selected harmonics of the returned signal are processed to form the ultrasound image. Briefly and in general terms, harmonic generation by the internal structures in the body is at least partially determined by the properties of the tissue that reflect the ultrasound energy, so that the presence of harmonics in the received echo may be used to generate useful information in the ultrasound image, as discussed in further detail in A. Bouakaz, E. Merks, C. Lancee, N. Bom, “Noninvasive Bladder Volume Measurements Based on Nonlinear Wave Distortions,” Ultrasound in Medicine & Biology, 30:4, pp. 469-476, which publication is incorporated by reference herein.

In selected ultrasound imaging applications, it is often desirable to distinguish between a bodily fluid and an adjacent tissue, or between bodily fluids of different types, such as between blood and various other bodily fluids. For example, when a selected anatomical portion is imaged using B-mode ultrasound imaging, a bodily fluid and certain adjacent soft tissues within the selected portion may be relatively indistinguishable in the resulting image. Moreover, when blood within the selected portion and other bodily fluids are imaged using B-mode ultrasound, images may be generated that similarly fail to properly distinguish the blood from the other bodily fluids.

Accordingly, a new imaging system is needed that permits the diagnostician to easily distinguish or discriminate between different fluid compositions, or between a bodily fluid and tissue. It is further desirable to render more easily detectable in an ultrasound image any boundaries between cavities that contain fluids of different composition, and between a bodily fluid and a bodily tissue.**

It is well known that bladder dysfunction is associated with a number of clinical conditions requiring treatment. In many of these cases it is important to accurately determine the volume of the bladder. Under other conditions such as post-operative recovery, where there is temporary loss of bladder sensation and/or loss of the normal voiding mechanism too much distention of the bladder has to be avoided. Under those conditions voiding by catheter introduction is carried out. However, serious disadvantages to unnecessary catheterization range from. the uncomfortable situation for the patient to serious possibilities of infection. Thus, a non-invasive quick measurement of bladder volume, with the patient usually in the supine position, is indicated. Sometimes the accurate determination of volume is indicated; sometimes however an indication is sufficient. Questions that may be asked are for instance: after voiding:” is there still too much urine left?”; or after surgery” is the bladder filling above a certain level so that voiding is necessary?”

Non-invasive procedures for bladder volume estimation are known, but are either unreliable or expensive or have some other significant disadvantages. Palpation and auscultatory percussion are known to be unreliable, while radiography and dye-excretion techniques are known to be similarly inaccurate. For assessing bladder volume, catheterization remains the “gold standard”. However, it is invasive, painful and might produce traumas or infections.

SUBJECT

The described technique concerns measurement of urine volume in the human bladder with the use of pulsed ultrasound with a limited number of ultrasound transducers.

In a first version a limited number of transducers are mounted in a transducer assembly. The assembly is positioned non-invasively at the body skin over the position of the bladder with the patient in a supine position. For acoustic contact a coupling gel may be used. Each ultrasound transducer in the assembly transmits and receives the ultrasound signal in a narrow beam through the contact plane. During the measurement the transducers are used in a certain succession. All transducers have been mounted in the assembly such that in transmission and reception successively the beams penetrate the area of the bladder in approximately the sagittal cross sectional plane. The sagittal plane is here defined as ANTERO-POSTERIOR plane of the body. One transducer beam direction is dorsal with in addition at least one transducer beam in the dorsal-caudal and one transducer beam in the dorsal-cranial direction. The volume is calculated on the basis of two bladder measurements defined in the sagittal plane as Depth (D) and Height (H). These measurements are derived on the basis of echo travel time from echoes originating at the anterior and posterior bladder wall. Depth is in principle a measurement in dorsal direction. Height is a measurement approximately in the cranial direction. The volume is calculated depending on the specific, filling dependent, measurement configuration following the formula D×H×K. Where K is an empirically measured, filling configuration dependant, correction factor. Beam directions and examples for D and H are illustrated in FIGS. 1 and 2.

In a second version of the described technique a single wide beam ultrasound transducer is positioned non-invasively at the body skin over the location of the bladder. The wide beam can be created by the curved surface of the transducer or by a flat acoustically active surface of for instance a disk shaped transducer supplied with a curved lens. Ultrasonic signals are transmitted and received in the wide, cone like, ultrasound beam and propagation is approximately spherical. Similar to the above described method a pulsed echo signal is transmitted at fundamental ultrasonic frequency. In this second version of the described technique echo data are analyzed as originating from a distance beyond the average position of the posterior (filled) bladder wall. The received echo signal will contain information over almost the entire bladder as encompassed by the wide ultrasound beam. Due to non-linearity, higher harmonic components will build up during propagation and thus be reflected in the returning echo.

Compared to propagation through normal tissue, the presence of higher harmonics in the signal is greatly stimulated when propagating through urine. Analyses of presence of higher harmonic components in relation to the fundamental frequency is used for indication of presence of urine in the bladder. Neutralizing patient variation as to obesity etc can also be accomplished by comparing echo signals received from sequentially transmitted pulses at low transmit power (linear propagation only) and pulse transmission at high power (enhancing non-linearity).

STATE OF THE ART

Non-invasive bladder volume measurement techniques with ultrasound echography have been described in the art. In principle, echography measures distance based on echo travel time. Early echo techniques did use a single ultrasound transducer and echo presentation was recorded as echo amplitude versus depth. West, K A: “Sonocystography: A method for measuring residual urine”, Scand J Urol Nephrol 1: pp 68-70, 1967 describes the subsequent use of some discrete beam directions. He does not have a separate transducer for each beam direction. His method is only qualitative, not instantaneous, and based on distance measurement to the dorsal posterior bladder wall. His method is not adjusted to specific, filling dependent, measuring configurations. A relation between the difference in echo travel time between echoes from the posterior an anterior bladder wall and the independently measured bladder volume has been reported by Holmes, J H: “Ultrasonic studies of the bladder”, J. Urology, Vol 97, pp. 654-663. His described volume measurement method is exclusively based on bladder depth measurement. Since the bladder changes in shape when filling, a single distance measurement is not precise enough to predict the entire bladder volume. No filling dependent measurement configuration is used.

Diagnostic ultrasound is today well known for real-time cross-sectional imaging of human organs. For cross-sectional imaging the sound beam has to be swept electronically or mechanically through the cross section to be imaged. Echoes are presented as intensity modulated dot on the display. The instruments are costly and require a skilled operator. Volume is sometimes calculated based on bladder contours obtained in two orthogonal planes with a geometric assumption of bladder shape. For 3-dimensional or volumetric echography the sound beam has to be swept through the entire organ. This further increases complexity, acquisition time of the data, and costs of the instrument.

HAKENBERG ET AL: “THE ESTIMATION OF BLADDER VOLUME BY SONOCYSTOGRAPHY”, J Urol, Vol 130, pp 249-251, have reported a simple method that is based on measuring the diameters obtained in a cross sectional image in the midline sagittal bladder plane only. The bladder volume has been related to bladder Height and Depth as follows: Volume is Height×Depth×6.6 ml. This formula showed a good correlation coefficient (r=0.942) with a relatively large average error of 30.1%. For this approach a two-dimensional imaging apparatus was required. The used apparatus is complex and is different from the method described in this application. It does not use a single wide beam transducer or a limited number of fixed transducers in an assembly or a combination of this.

An ultrasound apparatus for determining the bladder volume is shown in U.S. Pat. No. 4,926,871 in the name of Dipankar Ganguly et al. In this text, a number of possibilities are mentioned, amongst which a scan head embodiment referred to as a sparse linear array with transducers mounted at predetermined angles with sound beams pointing towards the same position. The volume is calculated according to a geometric model. In the claims an apparatus is described, involving an automatic calculation of bladder volume from ultrasound measurements in a first and second plane, which are substantially orthogonal to each other. Sound beams are deflected by a stepper motor. It requires a skilled operator to manipulate the scan head in a particular way to obtain the ultrasound measurements. For the volume calculation method described in this application no use is made of any geometrical model of the bladder, whereas only a limited number of sound beams approximately in the sagittal plane, or a single wide beam is used.

Volume measurement based on echographic sampling of the bladder with a hand guided transducer mounted in a panthograph has been described by Kruczkowski et al:” A non-invasive ultrasonic system to determine residual bladder volume”, IEEE Eng in Medicine & Biology Soc 10TH Ann Conf, pp 1623-1624. The sampling covers the entire bladder, follows a given pattern and is not limited to a single or two cross sections of the bladder. For the calculation he needs data from many beam directions. The acquisition procedure is time consuming and thus no instantaneous volume measurement results. The method described in this application is based on use of a single, wide beam or the use of a limited number of mutually fixed sound beams directions with instantaneous volume indication.

The hand steered transducer guiding for recording of echo data from the bladder has subsequently gained in acquisition speed by introduction of constructions whereby the transducer, and thus the beam, was mechanically swept. This nevertheless still requires an acquisition time equivalent to full acquisition procedure and thus does not yield an instantaneous display of volume. No instantaneous feedback on optimal positioning is thus available. An example of such methods is the BLADDERSCAN. In the Bladderscan Technology (registered trademark of Diagnostic Ultrasound Corporation) bladder volume is measured by interrogating a three-dimensional region containing the bladder and then performing image detection on the ultrasound signals returned from the region insonated. The three dimensional scan is achieved by performing twelve planar scans rotated by mechanically sweeping a transducer through a 97 degree arc in steps of 1.9 degrees. The three dimensional scanning requirement makes this instrument complex. It can not be compared with the simple approach described in this application.

Yet another ultrasound method “System for estimating bladder volume” is described by Ganguly et al in U.S. Pat. No. 5,964,710 dated Oct. 12, 1999. This method is based on bladder wall contour detection with echographically obtained data in a plurality of planes which subdivide the bladder. In each single plane of the plurality of planes a number of N transducers are positioned on a line to produce N ultrasound beams to measure at N positions the distance from front to back wall in the selected plan. From this the surface is derived. This procedure is repeated in the other planes as well. The volume is calculated from the weighted sum of the plurality of planes. In Ganguly's method the entire border of the bladder is echographically sampled in 3 dimensions. His method differs strongly from the method described in this application whereby only a single wide beam is used or a limited number of mutually fixed sound directions are used in approximately a sagittal plane with a filling dependent measurement configuration.

U.S. Pat. No. 6,359,190 describes a device for measuring the volume of a body cavity, such as a bladder or rectum, using ultrasound. The device is strapped to the body or incorporated into a garment such as a nappy or trainer pant. The device includes several transducers each aimed at a different region of the subject's bladder (a) to ensure that at least one ultrasound beam crosses the bladder despite variations in the way that the device has been positioned on the body, and (b) to enable the transducer with the strongest signal output to be used. An alarm signal may be output when the bladder reaches a predetermined threshold volume.

An important parameter for assessing bladder volume if this volume has to be derived from a limited number of beams or planes is the knowledge of bladder shape and position which can drastically vary with age, gender, filling degree and disease. In the adult patient the empty bladder has the shape of a triangular prism and is located behind the pubis. When it is progressively filled, there is first a distention of the bladder depth followed by an expansion of the bladder height. The bladder shape is complex and can not be represented by a single geometrical formula such as ellipsoid, sphere etc. This explains the large error that several studies obtained when a single geometric model was used. However there exists a correlation between the bladder height and the bladder widening with progressive filling.

In the first approach of the present invention an instrument is described which allows assessment of bladder volume by using only a few ultrasound beams appropriately oriented in approximately the sagittal plane. The narrow sound beams in principle diverge relative to each other. This allows covering a wide range of filling degrees of the bladder, from almost empty, when the bladder is located behind the pubis, to a full bladder that causes a substantial bladder height (See FIGS. 1 and 2). From each beam can be established, by detection of the posterior bladder wall echo, if this beam does pass a filled bladder. From the knowledge of all beams that do pass the filled bladder the appropriate filling or measurement configuration follows. The acoustic beams are positioned in such a way that the Depth D and Height H of the bladder can be estimated for the specific measurement configuration. The volume of urine is then computed from an empirical formula D×H×K that does not depend on any geometric model. K is a known, empirically established correction factor which is specific for each measurement configuration and has been established by calibrated bladder measurements on a prior series of patients. The accuracy of the first approach is thus based on an a prior known correction factor which is related to a specific filling degree, which in turn depends on the number of beams that intercept the filled bladder.

A second version of the instrument is based on the measurement of the presence of higher harmonics in the echo signal. For this approach the echo signal from a depth greater than the distance from the transducer to the posterior bladder wall must be analyzed. For a filled bladder in adults in a supine position, this depth W would be approximately 12 cm.

It is known that when sound pulses are transmitted at a fundamental frequency, higher harmonics of this fundamental frequency may be present in the received echographic signal. Non-linear distortion increases with distance, insonifying ultrasound energy and frequency. Attenuation diminishes the ultrasound amplitude with increasing propagation distance and reduces the higher harmonic energy. Since attenuation of the ultrasound signal in urine is low compared to tissue and non-linear distortion in urine is large compared to tissue it results that urine is very different from tissue in its ability to generate higher harmonics. We have measured the presence of higher harmonics in the echo signal from 12 cm depth when the bladder was filled. With an empty bladder the echoes obtained from the same depth did not contain higher harmonics.

The interest of higher harmonic signals in the ultrasound technique stems from echo contrast technology. Echo contrast material contains coated gas containing micro bubbles suspended in a fluid. These bubbles can create higher harmonic components in the echo signal due to non-linearity. This is used to indicate presence of contrast on the diagnostic image. A wide variety of pulse techniques is used to stimulate echographic visibility of contrast. These include multi pulse procedures, multi frequency procedures, power Doppler imaging, pulse coding, pulse inversion and other imaging methods. A survey is documented in “Ultrasound Contrast Agents” ISBN 1-85317-858-4 chapter 3 “Contrast-specific imaging methods” by de Jong et al. With a single transducer with wide sound beam, such as results with a curved acoustic element or a flat, disk shaped transducer plus curved lens, the propagating sound beam would encompass the entire bladder. The transducer must be designed to optimally transmit the fundamental ultrasound frequency and at the same time be capable to receive fundamental and higher harmonic echo signals. Broadband piezo-electric ceramic transducers have been described as well as combination transducers using ceramic in transmission and PVDF material in reception. In transmission a single or multi pulse procedure can be followed. If the returned echo signal with such a method would, in relation to the fundamental echo signal, be analyzed for the presence of higher harmonics, the presence of a certain level of bladder filling or the volume of urine can be established.

EP 0271214 describes an ultrasonic device for monitoring the volume of fluid in the human bladder by using reflected ultrasound signals to determine not only the position of the bladder back wall but also energy returned from the bladder back wall. EP '214 proposes that after bladder filling to approximately 60% capacity, the distance between the back wall and the front wall of the bladder stops increasing. However, additional reverberation in the back wall provides an increase in energy in the reflected signal which can be used to determine further increases in bladder volume.

SUMMARY

The present invention comprises ultrasound imaging systems and methods. In one aspect, an ultrasonography method includes creating a database that is representative of a tissue, a fluid, or a cavity of a body, and transmitting ultrasound pulses into a region-of-interest in a patient. Echoes are received from the region of interest, and based upon the received echoes, compiling an ultrasonic pattern of the region-of-interest is compiled. The pattern is processed by comparing the region-of-interest patterns to the pattern information stored in the database. A composition within the region-of-interest of the patient is then determined.

BRIEF DESCRIPTION OF THE DRAWINGS

The embodiments of the present invention are described in detail below with reference to the following drawings.

FIG. 1 is a side elevational view of a microprocessor-controlled transceiver according to an embodiment of the invention;

FIG. 2 is a representation of an ultrasound scan cone emanating from the transceiver in a conic shape formed by a plurality of three-dimensional distributed scan lines;

FIG. 3A is a representation of an ultrasound scan cone emanating from the transceiver in a conic shape formed by a rotational array of two-dimensional scan planes;

FIG. 3B is a representation of a scan plane of the rotational array;

FIG. 4 is a is depiction of the hand-held transceiver in use for scanning the abdominal area of a patient;

FIG. 5 is a perspective view of the hand-held transceiver device sitting in a communication cradle;

FIG. 5B illustrates a schematic view of an imaging system;

FIG. 6 depicts a schematic view of a plurality of transceivers in connection with a server;

FIG. 7 is a schematic view of a plurality of bladder wall measuring systems connected to a server over the Internet or other network;

FIG. 8 illustrates a scan plane image with diagrammatic scan lines overlaid on the image;

FIG. 9 is a schematic illustration of a scan line passing through a full bladder and a uterus;

FIG. 10 is a plot of echo amplitude versus scan line depth of the FIG. 9 schematic;

FIG. 11 is a schematic illustration of a scan line passing through a near empty bladder and the uterus;

FIG. 12 is a plot of echo amplitude versus scan line depth of the FIG.

FIG. 13 is a schematic illustration of a scan line passing through body cavities and surrounding tissues;

FIG. 14 is another spectral plot of a window function processed insonified region of FIG. 13 for a non-pregnant female subject having homogenous uterine fluid;

FIG. 15 is another spectral plot of a window function processed insonified region of FIG. 13 for a non-pregnant female subject having heterogeneous uterine fluid;

FIG. 16 is a calibration plot of harmonic power as a function of the bladder volume;

FIG. 17 is a calibration plot of harmonic ratio as a function of blood composition in amniotic fluid;

FIG. 18 is a method overview flowchart;

FIG. 19 is an expansion of sub-algorithm 206 of FIG. 18;

FIG. 20 is a schematic representation of four surface patch elements; and

FIG. 21 is a schematic representation of three scan lines passing through the subserosal and submucosal wall locations of an organ.

FIG. 1 Illustrates a sagittal (anteroposterior) cross sectional plane of a patient in supine position where a transducer assembly 1 with transducers A, B, C, D and E, is positioned on the abdominal wall just above the Symphysis Pubis 2 and the ultrasound beams are indicated to cross the area of the partially filled bladder 3. From the transducer assembly, the sound beam A intercepts the bladder area in dorso-caudal direction, soundbeam B intercepts the bladder in dorsal direction and sound beams C, D, and E respectively in dorso-cranial direction. In FIG. 1 the patient's leg is indicated by 4.

FIG. 2 Illustrates various bladder filling stages from an almost empty bladder to a strongly filled bladder and the corresponding measurement configurations. Depth D and Height H have been defined for each filling situation as indicated and are calculated from detected bladder wall echoes taking the specific measurement configuration into account. For each measurement configuration a specific Depth D and Height H is defined.

FIG. 3. Illustrates, by way of example for a transducer assembly with five transducers (here only A and D, necessary for calculation of H are shown), the calculation of Height H (5) in the measurement configuration when bladder posterior wall echoes are detected originating from sound beam A, B, C and D. This is the “filled bladder” measurement configuration shown in FIG. 2. Apparently no posterior wall echoes are detected in sound beam E because the bladder filling is not yet in a strongly filled stage and thus beam E does not intercept the bladder. Depth D is derived from beam B (not shown in FIG. 3).

FIG. 4 Represents a flow chart of the actions of the principal hardware components. In this block diagram a “useful” transducer signal occurs when bladder wall echoes are detectable in its sound beam.

FIG. 5. Illustrates a top view of five disk shaped transducers in a possible transducer assembly. The distance between transducers B, D and C, A, E and their positioning is such that all sound beams can be assumed to be in approximately a sagittal cross section through the bladder. Yet another transducer assembly with 4 transducers in a row is also illustrated.

FIG. 6. Illustrates a cross sectional view showing in the length direction a possible transducer and related sound beam orientation when five single transducers are used.

FIG. 7 Illustrates the sagittal cross sectional plane with a single wide beam transducer non-invasively positioned on the abdominal skin surface over the filled bladder 3. Echo signal is received from a range at depth W.

FIG. 8. Is a flow chart illustrating the principal steps taken by the bladder volume measurement instrument based on a single ultrasound wide beam where detection of presence of higher harmonics in the received signal from a give depth range is used to measure volume. Two different transmit levels are used to enhance the bladder effect and eliminate patient variation.

FIG. 9. Illustrates the measured received scattered power in the fundamental frequency to and the higher harmonic frequencies 2f₀ and 3f₀ in a situation with an empty versus a filled bladder.

FIG. 10 shows two possible transmit pulse sequences to enhance the difference between linear and non-linear sound propagation.

FIG. 11 Illustrates a possible look-up table based on prior calibrated patient bladder volume measurements relating presence of harmonic power in the received echo signal versus volume.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention relates to the ultrasound imaging of tissues and/or fluid-filled cavities having linear or non-linear acoustic properties. Many specific details of certain embodiments of the invention are set forth in the following description and in FIGS. 1 through 21 to provide a thorough understanding of such embodiments. One skilled in the art, however, will understand that the present invention may have additional embodiments, or that the present invention may be practiced without several of the details described in the following description.

FIG. 1 is a side elevational view of an ultrasound transceiver 10 according to an embodiment of the invention. The transceiver 10 includes a transceiver housing 18 having an outwardly extending handle 12 that is suitably configured to allow a user to manually manipulate the transceiver 10. The handle 12 includes a trigger 14 that allows the user to initiate an ultrasound scan of a selected anatomical portion, and a cavity selector 16, which will be described in greater detail below. The transceiver 10 also includes a transceiver dome 20 that contacts a surface portion of the patient when the selected anatomical portion is scanned, and a display 24 operable to view processed results from the ultrasound scan, and to allow operational interaction between the user and the transceiver 10. Accordingly, the display 24 may be configured to display alphanumeric data that indicates a proper and/or optimal position of the transceiver 10 relative to the selected anatomical portion. In other embodiments, two- or three-dimensional images of the selected anatomical region may be displayed on the display 24. The display 24 may be a liquid crystal display (LCD), a light emitting diode (LED) display, a cathode ray tube (CRT) display, or other suitable display devices operable to present alphanumeric data and/or graphical images to a user.

The transceiver 10 further includes a microprocessor (not shown in FIG. 1) and computational algorithms (also not shown in FIG. 1) that cooperatively provide enhanced ultrasound harmonic imaging that permits boundaries between different fluid compositions to be distinguished. In addition, the transceiver 10 may be suitably configured to distinguish between a bodily fluid and tissue, between dissimilar tissues, and/or between dissimilar bodily organs. The computational algorithms will be discussed in greater detail below. The transceiver 10 may also be coupled to a computer system (not shown in FIG. 1) that is operable to receive either digital or analog signals from the transceiver 10 and to process the signals to generate a desired ultrasound image. In addition, the computer system may also at least partially control the operation of the transceiver 10. The computer system may comprise any microprocessor-based computer or other computer systems, such as a mainframe that is capable of executing operating instructions and manipulating data. Accordingly, the computer system is not limited to a typical desktop or laptop computer device.

The operation of the transceiver 10 will now be described. The transceiver dome 20 of the transceiver 10 is positioned against a surface portion of a patient that is proximate to the anatomical portion to be scanned. The user then actuates the transceiver 10 by depressing the trigger 14. In response, the transceiver 10 transmits ultrasound signals into the body, and receives corresponding return echo signals that are at least partially processed by the transceiver 10 to generate an ultrasound image of the selected anatomical portion. In a particular embodiment, the transceiver 10 transmits ultrasound signals in a range that extends from approximately about two megahertz (MHz) to approximately about ten MHz.

Still referring to FIG. 1, the cavity selector 16 is structured to adjustably control the transmission and reception of ultrasound signals to the anatomy of a patient. In particular, the cavity selector 16 adapts the transceiver 10 to accommodate various anatomical details of male and female patients. For example, when the cavity selector 16 is adjusted to accommodate a male patient, the transceiver 10 is suitably configured to locate a single cavity, such as a urinary bladder in the male patient. In contrast, when the cavity selector 16 is adjusted to accommodate a female patient, the transceiver 10 is configured to image an anatomical portion having multiple cavities, such as a bodily region that includes a bladder and a uterus. Alternate embodiments of the transceiver 10 may include a cavity selector 16 that is configured to select a single cavity scanning mode, or a multiple cavity-scanning mode that may be used with male and/or female patients. The cavity selector 16 may thus permit a single cavity region to be imaged, or a multiple cavity region, such as a region that includes a lung and a heart to be imaged.

FIG. 2 is a representation of an ultrasound scan cone 30 emanating from the transceiver 10 of FIG. 1 that will be used to further describe the operation of the transceiver 10. The scan cone 30 has a substantially conic shape formed by a plurality of three-dimensional distributed scan lines. A scan cone 30 is shown emanating from the dome 20 of the transceiver 10 in encompassing a plurality of three-dimensional-distributed scan lines 31A-34E. The plurality of scan lines 31A-34E represent a line array in three-dimensional space. The scan lines within the line array are one-dimensional ultrasound A-lines that emanate from the transceiver 10 at different coordinate directions that taken as an aggregate form the scan cone 30. The different coordinate directions comprise a length r of a given scan line, and a rotational angle θ and a tilt angle φ. Thus, one or more points P along a scan line within the line array 31A-34E are defined by the distance r and the angular coordinates φ, and θ.

The plurality of three-dimensional distributed scan lines 31A-34E comprises a plurality of peripheral scan lines 31A-E and a plurality of internal scan lines 34A-D. The three-dimensional-distributed A-lines (scan lines) are not necessarily confined within a scan plane, but instead are directed to sweep throughout the internal regions and along the periphery of the scan cone 30. The three-dimensional-distributed scan lines not only would occupy a given scan plane in a three-dimensional array of two-dimensional scan planes, but also the inter-scan plane spaces, from the conic axis to and including the conic periphery. For example, assume line 34 B is a conical axis line and lines 31C and 34A are coplanar with line 34B. Lines 34A and 34B are separated by a tilt angle φ₁ and lines 31C and 34A are separated by a tilt angle φ₂. Similarly, lines 31F and 31C. are separated by a rotational angle θ₁ and lines 31D and 34C are separated by a rotational angle θ₂.

The internal scan lines are represented by scan lines 34A-C. The number and location of the internal scan lines emanating from the transceiver 10 is variable and may be selected to sufficiently visualize structures within the scan cone 30. The internal scan lines are not peripheral scan lines. The peripheral scan lines 31A-F occupy the conic periphery and converge near the apex of the scan cone 30, thus representing the peripheral limits of the scan cone 30.

FIG. 3A is a representation of an ultrasound scan cone emanating from the transceiver in a conic shape formed by a rotational array of two-dimensional scan planes. The scan cone 40 emanating from the dome 20 includes a plurality of scan planes 42 assembled as a rotational array. The scan planes within the rotational array are angularly separated by an angle θ.

FIG. 3B is a representation of the scan plane of the rotational array. A scan plane 42 is formed by a scan line 48 that rotationally pivots between a first leg 44 and a second leg 46 about a pivot angle φ. The depth of the scan plane 42 is determined by the effective length r of the scan line 48. The area of the scan plane 42 is determined as a product of the length r of the scan line 48 and region swept by the scan line 48 as it migrates about pivot angle φbetween the first and second legs 44 and 46.

FIG. 4 is an isometric view of the transceiver 10 positioned on an external portion of a patient 26 that will be used to describe a method of data acquisition for identifying fluids in bodily cavities according to an embodiment of the invention. The transceiver 10 is positioned against a surface portion of the patient 26, and a targeting phase is initiated. In a particular embodiment, the transceiver 10 is then operated in a two-dimensional continuous acquisition mode, which permits data to be continuously acquired and presented as a discrete scan plane image on the display 24 (or another external display device) as the operator physically moves the transceiver 10 across various external portions of the patient 26. In this embodiment, the operator moves the transceiver 10 around an abdominal region and depresses the trigger 14 of the transceiver 10 to continuously acquire real-time two-dimensional images that may be continuously viewed on the display 24, or on another display device. For example, when an anatomical portion that contains a urinary bladder is imaged, urine confined within the bladder appears as a dark region, and a urine fluid area may be calculated. An alphanumeric indication of the urine fluid area (for example, in cm²) may also be calculated and visually presented on the display 24. Similarly, if the patient 26 is a pregnant female, amniotic fluid within the uterus may also be imaged and a corresponding amniotic fluid area may be calculated and displayed on the display 24. After acquisition of the two dimensional measurements, the volume of urine and amniotic fluids are measured in the respective bladder and uterus by acquiring a 3-D scan as a multiple scan plane array similar to the scan cone 40. Alternatively, if the two-dimensional measurements are acquired as three-dimensional distributed scan lines, a three-dimensional scan is accomplished as a three-dimensional scan cone of three-dimensional distributed scan lines similar to the scan cone 30. A cavity selector 16 (as shown in FIG. 1) is engaged to detect and measure the volumes of either single or multiple cavities in a subject. In a particular embodiment where the transceiver 10 is positioned over the symphasis pubis for acquisition of three-dimensional ultrasound images, a single cavity includes one of the bladder and the uterus, and a multiple cavity includes the bladder and the uterus.

FIG. 5 is an isometric view of transceiver 10 according to another embodiment of the invention. The transceiver 10 is structured to be received by a support cradle 50. The support cradle 50 is coupled to a power supply 51 that provides electrical energy to the cradle 50 that is communicated to the transceiver 10 either conductively or inductively to provide a charging current to a power supply positioned within the transceiver 10. The support cradle 50 is also configured to receive ultrasound data from the transceiver 10 when positioned in the cradle 50, which may be transferred to an external processor (not shown in FIG. 3) through a digital communications link 38, such as a link that employs a universal serial bus (USB), a FIREWIRE bus in conformity with IEEE-1394, an RS-232 compatible link, or other similar communications links in conformity with still other protocols. In other embodiments, the communications link 38 may be a wireless link, such as wireless local area network (LAN) or a wireless wide area network (WAN). Alternatively, the cradle 50 may be powered by the link 38.

The communications link 38 may also advantageously provide a means for transferring imaging data from the transceiver 10, and for transferring software updates or software revisions from the external processor to the transceiver 10. In a particular embodiment, the cradle 50 may include a memory device operable to retain digital data received from the transceiver 10 before the data is transferred to the external processor through the communications link 38.

FIG. 5B illustrates a schematic view of an imaging system 55. The imaging system 55 includes the transceiver 10 positioned in the supporting cradle 50. The communications link 38 connects the transceiver 10 housed in the cradle 12 with a computer 62. The computer 62 may be a desktop, laptop, or other microprocessor-based portable computing device. Data from the transceiver 10 is routed through the cradle 50 to the computer 62 via the communications link 38. The communications link 38 may be a conductive link, as shown in FIG. 5B, or it may be a wireless radio frequency link or an optical link, such as a wireless infrared link. Within the computer 62 are executable programs to implement the algorithms of the particular embodiments, including the processing of ultrasound signals, retrieving imaging programs, and instructions to perform ultrasound enhancement procedures. Various ultrasound images are developed by processing the ultrasound signal data, including one-dimensional ultrasound images, two-dimensional images, three-dimensional renderings, and enhanced images from the retrieved imaging programs and instructions. The generated images may be stored within the computer 62.

FIG. 6 is a partial isometric and diagrammatic view of a networked imaging system 60 according to another embodiment of the invention. The imaging system 60 includes one or more transceivers 10 in accordance with one or more of the previously disclosed embodiments. The one or more transceivers 10 may be positioned in supporting cradles 50 that are operably coupled to portable computing devices 62, which in turn, are suitably configured to receive imaging data from the one or more transceivers 10 through the communications link 38. The communications link 38 may be a conductive link, as shown in FIG. 6, or it may be a wireless radio frequency link or an optical link, such as a wireless infrared link. The portable computing devices 62 communicate with a server 66 over a communications network 72. Although two transceivers 10 are shown in the networked imaging system 60 shown in FIG. 6, fewer than two, or more than two transceivers 10 may be present. In addition, the processing of ultrasound signals may be divided between the transceiver 10, the portable computing devices 62, and the server 66. For example, the transceiver 10 may be configured to process the ultrasound signals and generate an ultrasound image using algorithms in accordance with other embodiments of the invention, or alternately, the ultrasound image may be generated by the portable computing device 62 or even by the server 66 after receiving ultrasound signals from the transceiver 10. In a particular embodiment of the networked imaging system 60, the imaging algorithms that generate the enhanced ultrasound images reside on the server 66. Each of the portable computing devices 62 accordingly receives signals acquired from the transceivers 10 through the cradles 50 and stores the signals in the portable computing device 62. The computing device 62 subsequently retrieves imaging programs and instructions to perform the additional ultrasound enhancement procedures from the server 66. Thereafter, each personal computing device 62 generates various ultrasound images by processing the ultrasound data, including one-dimensional ultrasound images, two-dimensional images, three-dimensional renderings, and enhanced images from the retrieved imaging programs and instructions. The generated images may be stored on the server 66.

In another particular embodiment, the imaging programs and the instructions reside exclusively on the server 50 and are executed on the server 50. Each portable computing device 62 receives the acquired signals from the transceivers 10 through the cradle 50 and transfers the acquired signals to the portable computing device 62. The device 62 subsequently communicates the signals to the server 66 and processes the signals to generate the desired ultrasound images, including one-dimensional images, two-dimensional images, three-dimensional renderings, and other similar images. The ultrasound images may be stored on the server 66, or alternately, the images may be transferred to one or more of the personal computing devices 62.

FIG. 7 is a partial isometric and diagrammatic view of a networked imaging system 80 according to still another embodiment of the invention. Many of the elements of the present embodiment have been discussed in detail in connection with other embodiments, and in the interest of brevity, will not be discussed further. The networked imaging system 80 includes a public data network 82 interposed between the communications network 72 and the server 66. The public data network 82 may include a LAN, a WAN, or the Internet. Accordingly, other computational devices 84 associated with the public data network 82 may communicate imaging data and/or ultrasound images with the portable computing devices 62 and the server 66. Although two transceivers 10 are shown in the networked imaging system 80 shown in FIG. 7, fewer than two, or more than two transceivers 10 may be present.

FIG. 8 is an ultrasound image 90 of a bodily portion of a patient that will be used to describe other embodiments of the invention. The image 90 is formed by projecting a plurality of scan lines 48 downwardly into a selected anatomical portion of the patient to form the fan-like scan plane 42. The scan plane 42 may be rotated about an axis that extends through the transceiver 10 to generate a scan cone 40 (as shown in FIG. 3A) to obtain three-dimensional imaging information for the selected anatomical portion. Accordingly, when ultrasound energy is projected into the selected anatomical portion, various internal structures may reflect the ultrasound energy, including a bladder, a front bladder wall and a back bladder wall. The bladder may contain a volume of a fluid, such as urine, as shown in FIG. 8. The foregoing structures typically present imaging resolution difficulties. In particular, an ultrasound image may fail to adequately resolve a fluid-filled cavity, or a tissue that forms a boundary of the fluid-filled cavity, or still other structural details present in the imaged anatomical portion. Moreover, the foregoing structures generally respond to ultrasound energy in a non-linear manner, so that reflected ultrasound echoes include one or more harmonics of a fundamental ultrasound frequency.

One measure of the non-linear behavior of various fluids and tissues is the Goldberg number (G). G is a dimensionless quantity that generally relates ultrasound attenuation to harmonic distortion due to non-linear effects in a tissue or a fluid when subjected to ultrasound energy. Accordingly, when G is about one, non-linear effects are comparable to attenuation effects in the tissue. When G is much greater than one, such as for water or urine, the nonlinear processes are dominant. When G is less than one, as in soft tissue, attenuation effects are more dominant. For example, it is known that fatty tissue has a Goldberg number of approximately about 0.27, while blood, liver, and muscle have a G value of approximately about one. In contrast, fluids such as urine have a G value of approximately about 104.

With reference now also to FIG. 9, the scan plane 42 may include at least one scan line 102 that extends through a bladder and a uterus of a female patient. In this condition, referred to as a “case 1” condition in FIG. 9; the bladder includes a relatively large volume of urine. Typically, the bladder and the uterus appear as low echogenicity regions when the anatomical region shown in FIG. 9 is scanned. Known image processing software (incorporated herein by reference from one or more of the references listed in the priority claim section) may be used to image the shallowest region of low echogenicity. Since the low echogenicity region is generally preferentially selected, so that no imaging ambiguity exists, and the bladder is therefore readily identifiable.

Referring now to FIG. 10, a typical echo amplitude response for the anatomical region of FIG. 9 is shown. The echo amplitude response as depicted in FIG. 10 may be obtained through application of one or more of the algorithms incorporated by reference herein. For example, the computational algorithms disclosed in U.S. Pat. No. 6,676,605 to Barnard et al, U.S. Pat. No. 5,235,985 to McMorrow et al, and U.S. Pat. No. 4,926,871 to Ganguly et al, may be used.

When both organs are relatively filled with a fluid (as shown in FIG. 9), the edges of the bladder and uterus are relatively detectable and are thus generally distinguishable. In this case, the relatively full bladder presents a relatively U-shaped valley at a shallower bodily depth. In contrast, a corresponding U-shaped plateau presented by the uterus is generally identifiable at a greater bodily depth. Thus, while the embodiments of the present invention can improve accuracy and diagnosis in the foregoing situation where both organs are relatively filled with a bodily fluid, the embodiments are also suited to imaging bodily regions when one or both of the organs is less than full.

FIG. 11 is a diagrammatic representation of the anatomical region of FIG. 9, wherein the at least one scan line 104 extends through the bladder and the uterus when the bladder contains a relatively low volume of urine. This condition is referred to as “case 2” in FIG. 11. Because the bladder volume is greatly reduced in the Case 2 situation in comparison to the Case 1 condition shown in FIG. 9, the low echogenicity region is now principally located in the uterus. In current image processing methods, an average of the low echogenicity regions is compared to a threshold value to distinguish between the bladder and the uterus, which may tend to contribute to imaging errors. FIG. 12 is an echo amplitude response corresponding to the anatomical region of FIG. 11. The relatively empty bladder presents a relatively narrow valley. In contrast, the uterus generates a relatively wider U-shaped valley. As a consequence, the bladder is less readily distinguishable from the uterus when the bladder contains a low volume of urine. The disclosed embodiments better address the foregoing problems by using algorithms that more accurately detect ultrasound signal harmonic differences between cavity-residing fluids adjacent to enclosing tissue interfaces.

FIG. 13 is a diagrammatic representation that will be used to illustrate a method for imaging an anatomical region according to an embodiment of the invention. An A-mode scan line 106 is projected into a first bodily cavity 110, for example, a bladder and a uterus as depicted in FIGS. 9 and 11 for the respective case 1 (which corresponds to a relatively full bladder) and case 2 (which corresponds to a nearly empty bladder). The projected ultrasonic waveform is altered by the tissue along the distance d₁ of the scan line 106, which corresponds to tissue preceding the cavity 110, and by the distance d₂ that corresponds to an interior portion of the cavity 110 along the scan line 106. Other intervening cavities along the scan line 106, such as a uterus, may also alter the projected ultrasound waveform.

The first cavity 110 is hypoechoic and designated as region R_H. Other differentially echoic regions illustrated in the scan plane 42 include hypoechoic regions R₃, R₄, and R₅. Ultrasound energy passing through and within the body cavities 110 along scan line 106 may be subjected to an image analysis algorithm to determine respective volumes of each cavity, namely VI corresponding to the cavity 110. Signals reflected from the back wall or other boundaries of the first body cavity 110 are window function processed in a window region designated as WR1. The WR1 region spans a portion of the tissue adjacent and distal to the cavity 110 backwall interface and further spans a portion of the cavity space along the scan line 106 proximate to the cavity 110-backwall interfaces.

In the WRI region, window function processing determines the raw data comprising the fundamental frequency f₀ and a selected higher order harmonic 2f₀ that is generated within the WRI space along the scan line 106. The magnitude of the higher order harmonic generated within WRI varies because different tissues and/or fluids are encountered by the scan line 106 as it projects into the body. Consequently, a fluid volume and a fluid composition within the cavity 110 alters the magnitude of the higher order harmonic 2f₀ near the scan line 106 that is proximate to the back wall interface of the cavity 110.

FIG. 14 is a spectral plot of the insonified region of FIG. 13 that corresponds to a non-pregnant female with a uterus and nearly full bladder. The fundamental frequency f₀ has a peak value 140 and the higher order harmonic 2f₀ has a peak value 142. The fundamental and harmonic peaks 140 and 142 are a result of window function processing the corresponding echo amplitude response. The magnitude of the harmonic peak 142 may be normalized by dividing the peak 142 by the fundamental frequency peak 140. Accordingly, it is noted that a high Goldberg number stemming from urine in the nearly full bladder corresponds to a high magnitude for the frequency ratio. Different urine volumes and/or the presence of other organs, such as a uterus may also alter the magnitude of the frequency ratio. The magnitude of the second harmonic peak 142 in the first cavity 110 is affected by the presence of the uterus and the urine volume and urine composition contained within the bladder. The composition and volume of the urine may thus be determined.

FIG. 15 is a spectral plot of the insonified region of FIG. 13 in a non-pregnant female with a uterus and nearly empty bladder and having greater amounts of blood and tissue containing uterine fluid. The spectral plot is within the window region WRI and shows the fundamental peak 150 and the higher order harmonic peak 152 resulting from frequency domain processing of the corresponding echo response. As shown, the harmonic spectrum within WRI of the non-pregnant female exhibits a generally lower Goldberg number than a non-pregnant female with a substantially greater fluid volume. The magnitude of the frequency ratio (2f₀/f₀) is correspondingly lower. Urine fluids mixed with blood at variable compositions may also alter the magnitude of the higher order harmonic peak 152 shown in FIG. 15. Accordingly, the reflected spectral components generated within the first cavity 110 may have a still lower harmonic ratio (2f₀/f₀) as compared to unmixed uterine fluids depicted in FIG. 14, since the mixed fluid mixtures generally exhibit a lower Goldberg number.

With reference still to FIGS. 14 and 15, an embodiment of the invention will now be described. The present embodiment is based on non-linear wave propagation and variations in the attenuation of ultrasound energy in body fluids. The back wall ultrasound spectrum is processed to determine a reflected harmonic content, and this harmonic content is compared to the content of the fundamental ultrasound energy by forming the frequency ratio. The resulting value may then be adjusted for differences in attenuation at a selected frequency j between f₀ and 2f₀ in the intervening tissue i, as described below.

For a selected window, the total attenuation of the fundamental frequency component may be expressed as follows in equation E1:

A_f0=2d₁σ₁₁+2d₂σ₂₁≈2d₁σ₁₁ dB  E1

• • where: σ_ij=attenuation coefficient of a tissue i at a frequency j and distances d₁ and d₂ are as shown in FIG. 13.

While the total attenuation of the higher order harmonic frequency component may be expressed as follows in equation E2:

A_2f0=2d₁σ₁₂+2d₂σ₂₂≈2d₁σ₁₂ dB  E2

A difference in the attenuation of higher order harmonic component to the fundamental component is therefore defined by equation E3:

A_ratio=2d₁(σ₂₁−σ₁₁) dB.  E3

For example, in soft tissue having an attenuation factor of about 1.1 dB/cm, the attenuation coefficient σ₁₂ is approximately about 3.7 dB/cm, when f₀=3.7 MHz. Accordingly, the amplitude ratio becomes in equation E4:

A_ratio=2d₁(3.7 dB/cm)  E4

Based upon the foregoing, harmonic power amplitudes and frequency ratios may be derived and associated with known fluid volumes and fluid compositions derived from living subjects and/or non-living experimental devices, which are then encoded into readily accessible look-up tables, calibration plots or other suitable means for encoding data having utility in measuring the fluid volumes or identifying fluid compositions in an insonified subject.

FIG. 16 is an example of a calibration plot of harmonic power as a function of the bladder volume in a subject. The harmonic power may be obtained from a look-up table that includes data corresponding to different bodily tissues and fluids. The calibration plot thus permits the determination of a urine volume when the higher order amplitude and the fundamental frequency amplitudes are expressed as the ratio 2f₀/f₀. Alternate embodiments may include calibration plots for other fluid compositions that stem from data in respective look-up tables that are enhanced by the foregoing window functions. For example, a mixture of amniotic fluid and blood, or amniotic fluid/blood/urine mixtures would have a particular calibration plot. The plot shown in FIG. 16 may alternately be expressed using a suitable curve fitting procedure, such as, for example, a linear least squares procedure, a spline procedure, a polynomial procedure or other known curve fitting procedures.

In still another particular embodiment, the Goldberg number may be used to distinguish between a urinary bladder and a uterus in post-void scans in a female subject. Because the uterus generally appears as a dark structure in ultrasound images, it may be erroneously identified as the urinary bladder in post-void scans. To avoid this, the present embodiment provides that the harmonic amplitudes are calculated on a post-void scan of a female bladder. If a selected combination of harmonic amplitudes is higher than a selected threshold value, the scan likely contains a fluid. Otherwise, the scan likely contains the uterus of the female subject.

In still yet another particular embodiment, the harmonic amplitudes may be calculated based upon one or more selected ultrasound lines, or upon all of the ultrasound lines within the total image. In addition, the harmonics may be calculated within a region-of-interest or lines-of-interest when guided by features detected on the B-mode image, such as a region behind a posterior wall of a fluid-filled cavity.

In still another particular embodiment, inter-patient variability in the harmonic amplitudes may be normalized or otherwise adjusted using a combination of transmission features, such as, for example, frequency, peak-to-peak voltage, pulse length, and other transmission features. Other features may be extracted from B-mode images, such as a depth of an anterior wall of a fluid-filled cavity.

FIG. 17 is a calibration plot of a harmonic ratio expressed as a function of a blood composition in the amniotic fluid. A dashed calibration line may be used to graphically determine a blood percentage composition. The Goldberg number may also be used to differentiate between various bodily fluid compositions, such as, for example, a transudate and an exudate fluid developed in a lung of a subject, or in compositions of urine. It may also be used to detect the presence of blood within a bodily fluid, such as in amniotic fluid, or in urine. The harmonic ratio differences between the fluids may thus be used to identify a composition of a fluid. For example, as shown in FIG. 17, a harmonic ratio of 3 corresponds to an amniotic fluid having a blood composition of approximately 20%. Similarly, a harmonic ratio near 4.5 corresponds to an amniotic fluid having approximately 60% blood composition.

In still yet another particular embodiment, the Goldberg number may also be used to distinguish between blood and amniotic fluid in the uterus of a pregnant female. Because blood has a significantly lower Goldberg number compared to amniotic fluid, the second harmonic distortion resulting from a region containing blood is different from a region containing amniotic fluid. The Goldberg number and the harmonic ratio may thus be utilized to differentiate between blood and amniotic fluid and to confirm an identification of the fluid that is isonofied. For example, certain pregnant subjects having a low amniotic fluid volume. Accordingly, the uterus becomes more engorged with blood, making identification of the amniotic fluid regions more difficult. To address this problem, the embodiments of the present invention utilize the Goldberg number and harmonic ratio to assist in the identification process. In a further example, blood appears very dark and similar in appearance to amniotic fluid in a B-mode image. Thus, while measuring an amniotic fluid volume in B-mode images, blood may be detected in the umbilical cord or in the vessels in walls of the uterus and erroneously identified as amniotic fluid. Using the second harmonic distortion in conjunction with the Goldberg Number assists in discriminating between amniotic fluid and blood.

In still another particular embodiment, the Goldberg number may also be used to identify, classify, and measure a volume of a fluid in the lungs for pleural effusion.

In another embodiment, an ultrasound image may be created that shows selected combinations of harmonic amplitudes throughout the ultrasound image that permits detection of various fluid regions that would typically be indicated by presence of higher harmonics. The selected combination of the harmonic reflections may be embodied in a software program, or alternately as an improvement to an existing software program in conventional harmonic imaging ultrasound equipment. The harmonic ratio may then be normalized by the factor, A_ratio. The results may then be compared to an empirically derived look up table or calibration plot that describes one of the attenuation or harmonic characteristics for each kind of tissue (see FIG. 16). Accordingly, tissues and bodily fluids with low Goldberg numbers and low harmonic ratios can thus be differentiated from tissue and body fluids with higher Goldberg numbers and higher harmonic ratios. The use of the foregoing harmonic ratios and look-up tables (or equivalent calibration plots) in the manner described provides another basis to further differentiate and enhance a display of fluid and tissue regions. For example, a low volume to near empty bladder may be suitably differentiated from a uterine cavity and adjacent tissue.

In another embodiment of the invention, an additional adjustment may be performed to compensate for a “shock formation distance” that may occur along a given scan line. Shock formation distances relate to Goldberg numbers as a consequence of an energy transfer that occurs in the tissues and nearby fluids as the fundamental ultrasound frequency is transformed to a harmonic frequency. For example, an excitation frequency of 3.7 MHz results in significant harmonic generation in human tissue. Thus, the distance d₂ (FIG. 13) may also be used to adjust the harmonic amplitude ratio.

FIG. 18 illustrates a flow chart of a method 200 of measuring fluid volumes and classifying fluid compositions, according to an embodiment of the invention. The method 200 begins by creating a database that includes attenuation and/or harmonic characteristics of tissues, fluids, and cavities of a body at block 202. The database may be further characterized by sex, age, morphological, physiological, and pathological states. The method 200 continues by isonifying a selected region of a patient at block 204. Thereafter, ultrasonic patterns of the insonified region of the patient are compiled at block 206 (see FIG. 19 below). The method 200 continues by processing the ultrasonic patterns and comparing the processed patterns to the database information at block 208. For example, the processed patterns may be compared using the volume calibration plot of FIG. 16 and the composition calibration plot of FIG. 17. Thereafter, the method 200 concludes by determining at least one of a composition of the insonified region and a volume of the insonified region based upon the comparison of the patient's ultrasonic patterns to the database's content at block 210. Ultrasonic measurement data obtained from the subject at block 210 may then be applied to a volume calibration plot similar to FIG. 16 to obtain volume measurements. Similarly, ultrasonic measurement data obtained from the subject at block 210 can be applied to a composition calibration plot similar to FIG. 17 to obtain fluid composition measurements.

Still referring to FIG. 18, the processed patterns from the subject may be compared to the look-up table or calibration plot as depicted in FIG. 16 to obtain volume information. To obtain a compositional determination or classification of a given detected fluid, in accord with FIG. 17, the type of bodily fluids contained within a cavity may be ascertained through a comparative analysis of the Goldberg numbers, harmonic ratios, and attenuation factors within the insonified region. The Goldberg numbers, harmonic ratios, and attenuation factors stored within a database may then be accessed to determine the fluid composition. A fluid composition may thus be determined by accessing a calibration plot, interpolating from a look-up table, or applying regression analysis, as previously described. The volume and compositional look up tables or calibration plots illustrated in FIGS. 16 and 17 may be obtained from ultrasound information databases derived from simulated human models, accumulating clinical measurements obtained from patients stored in a separate database, or a combination of simulated models and clinical measurements. Other databases may include simulated animal models with or without a veterinary-based database. Yet other databases may include a combination of human and veterinary sourced databases.

Similarly, tissue types or combinations thereof within the insonified region are determined by comparative analysis between the Goldberg numbers, harmonic ratios, and attenuation factors presented by the insonified region of the patient to those same numbers, ratios, and factors stored in the database.

FIG. 19 is an expansion of sub-algorithm 206 of FIG. 18. Sub-algorithm 206 permits the determination of volume of an organ wall, the mass of an organ wall, the internal organ volume defined by an inner perimeter of an organ wall, and the outer organ volume defined by the outer perimeter of an organ wall from echogenic patterns received from an insonified region. The ultrasonic patterns of the insonified region having at least one organ of the patient are compiled at process block 206-2. Once the wall locations are identified, the wall locations, demodulated magnitude data, and a subset of quadrature amplitude demodulated signals in the region of the anterior bladder wall are directed to the microprocessor for further analysis according to the algorithm illustrated in FIG. 19 for the particular embodiments. First, ultrasound data is acquired relative to the bladder, uterus, or other organs as shown in the first block 206-2. In general, bladder-specific data can be acquired by a user who manipulates the transceiver 10 while viewing the received data on a display screen and then positioning the transceiver 10 as necessary so that an organ or organs, such as a bladder and uterus, are sufficiently within the field of view of the cone as depicted in FIGS. 2 and 3A.

Referring again to FIG. 19, and limiting the discussion to a specific organ, for example a bladder, echogenic data is collected by the transceiver 10. After obtaining ultrasound bladder data, the ultrasound data is processed to determine if the bladder contains approximately 200 to approximately 400 ml, as shown in the second process block 206-4 represented as a decision diamond. If “No” to the query “200 ml≦volume≦400 ml?”, then the bladder is allowed to accumulate approximately 200 to approximately 400 ml, as shown in the third process block 206-6, or, if “Yes”, meaning the bladder already contains the preferred approximate 200-400 ml volume, then the locations of the bladder walls, as shown in the fourth block 206-8, may be undertaken. The determination of organ wall locations and other such exterior boundaries within an ultrasound scan are within the capability of ultrasound devices presently on the market. In general, however, the process determines the length of a scan line from the transceiver dome to the bladder wall. The data, including wall locations, is stored in the memory of the computer 62 and is used to determine whether or not the bladder volume is within a range of approximately 200 to approximately 400 ml. If the bladder volume is within that range, the ultrasound data is used to determine the actual surface area from the wall locations, as indicated in the fifth block 206-10. The application of previously described methods using harmonic ratios, powers, and Goldberg G-numbers may be used to enhance the accuracy of thickness, area, volume, and mass determinations of bladders holding fluids within the approximate 200-400 ml range. The surface area calculation is explained with regard to FIG. 21 below and allows for calculation of an outer bladder wall surface area defined by subserosal locations 372A and 372B and an inner bladder wall surface area defined by submucosal locations 374A and 374B. While calculating the surface area in the fifth block 206-10, reflected ultrasound waves are received from the anterior bladder wall, as indicated in the sixth block 206-12. Although these tasks are preferably conducted in parallel, they may alternatively be processed in series. Thereafter, as shown in the seventh block 206-16, the bladder wall thickness is determined from the coherent signals that overlap at the wall locations. The determination of bladder wall thickness is explained in greater detail below. Finally, as shown in the seventh block 206-16, the bladder wall distance is computed as a difference between panterior and posterior submucosal bladder wall locations. Thereafter, at the eighth process block 206-20, the internal bladder volume is computed as a function of the internal bladder wall distances and the area of the internal bladder wall.

The volume restriction indicated in the previous paragraph is included as the range of bladder volumes that allow for an optimal measurement of the bladder wall mass, bladder wall volume, and internal bladder volumes. The volume and mass calculations may be performed at a volume not in this range, but will result in a less accurate measurement that can be corrected by application of the foregoing described methods using harmonic ratios, powers, and Goldberg G-numbers. For example, bladders having less than 200 ml or that are near empty, the foregoing described methods using harmonic ratios, powers, and Goldberg G-numbers will improve the accuracy of determining bladder wall thicknesses, volumes and mass, and internal and outer bladder volumes. For bladders having fluid volumes substantially greater than 400 ml, for example bladder volumes of 1000 ml to multi-liters, the invention will utilize scan lines greater than 20 cm to accommodate the larger bladder sizes. The invention may be applied to measure the thicknesses, masses, and volumes of internal organs of human and animals. The length of the scan lines is adjusted to match the dimension of the internal organ scanned.

The surface area measurement of fifth block 206-4 is performed by integrating the area of interpolating surface patch functions defined by the wall locations. The mathematical calculations are provided below in greater detail.

The surface of the bladder is defined to be S. This surface corresponds to the actual surface of the bladder determined by analysis of the wall locations of the bladder. Since this shape is not known in advance, modeling the bladder as a sphere or an ellipsoid provides only a crude approximation of the surface. Instead, the surface S is defined as a construction of a series of individual surface patches s_i,j, where i and j count through the latitude and longitude components of the surface, similar to the division of the Earth's surface into lines of latitude and longitude. The area of the bladder surface, S, is defined as the sum of all the individual surface patches, so that S=Σs_i,j.

FIG. 20 is a schematic representation of four surface patch elements. As depicted in three dimensions in FIG. 20, by way of example, five scan planes 320-328 are seen transmitted substantially longitudinally across a subserosal wall location 332 referenced to a tri-axis plotting grid 340. The five scan planes include the first scan plane 320, the second scan plane 322, the third scan plane 324, the fourth scan plane 326, and the fifth scan plane 328. The scan planes are represented in the preceding formulas as subscripted variable j. Substantially normal to the five longitudinal scan planes are five latitudinal integration lines 360-368 that include a first integration line 360, a second integration line 362, a third integration line 364, a fourth integration line 366, and a fifth integration line 368. The integration lines are represented in the preceding formulas as subscripted variable i.

By way of example, four surface patch functions are highlighted in FIG. 20 as the subserosal wall location 372. The i and j subscripts mentioned previously correspond to indices for the lines of latitude and longitude of the bladder surface. For the purposes of this discussion, i will correspond to lines of longitude and j will correspond to lines of latitude although it should be noted the meanings of i and j can be interchanged with a mathematically equivalent result. Using the scan plane and integration line definitions provided in FIG. 20, the four surface patch functions are identified, in the clockwise direction starting in the upper left, as s_322,362, s_324,362, s_324,364, and s_322,364.

The surface patches are defined as functions of the patch coordinates, s_i,j(u,v). The patch coordinates u and v, are defined such that 0<u, v<1 where 0 represents the starting latitude or longitude coordinate (the i and j locations), and 1 represents the next latitude or longitude coordinate (the i+1 and j+1 locations). The surface function could also be expressed in Cartesian coordinates where s_i,j(u,v)=x_i,j(u,v)i+y_i,j(u,v)j+z_i,j(u,v)k where i, j, k, are unit vectors in the x-, y-, and z-directions respectively. In vector form, the definition of a surface patch function is given in Equation 1. k, are unit vectors in the x-, y-, and z-directions respectively. In vector form, the definition of a surface patch function is given in equation E5.

$\begin{array}{cc}{s}_{i,j}\left(u,v\right)=\left[\begin{array}{c}{x}_{i,j}\left(u,v\right)\\ y_{i,j}\left(u,v\right)\\ z_{i,j}\left(u,v\right)\end{array}\right]& \mathrm{E5}\end{array}$

With the definitions of surface patch functions complete, attention can turn to the surface area calculation represented in the fifth block 206-10 of FIG. 20. The surface area of S, A(S), can be defined as the integration of an area element over the surface S, as shown in equation E6.

$\begin{array}{cc}A\left(S\right)={\int }_sA& \mathrm{E6}\end{array}$

Since S is composed of a number of the patch surface functions, the calculation for the area of the surface S can be rewritten as the sum of the areas of the individual surface patch functions as in equation E7.

$\begin{array}{cc}A\left(S\right)=\sum _{i,j}A\left(s_{i,j}\right).& \mathrm{E7}\end{array}$

Similarly to equation E5 for the entire surface, the area of the surface patch is the integration of an area element over the surface patch, shown in equation E8.

$\begin{array}{cc}A\left(S_{i,j}\right)={\int }_{s_{i,j}}A_{i,j}& \mathrm{E8}\end{array}$

The integration over the surface patch function can be simplified computationally by transforming the integration over the surface to a double integration over the patch coordinates u and v. The transformation between the surface integration and the patch coordinate integration is shown in equation E9.

$\begin{array}{cc}{\int }_{s_{i,j}}A_{i,j}={\int }_{u=0}^1{\int }_{v=0}^1\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}vu& \mathrm{E9}\end{array}$

By substituting Equation 5 into Equation 4, and Equation 4 into Equation 3, the area for the entire surface can be calculated. The result of these substitutions is shown in equation E10.

$\begin{array}{cc}A\left(S\right)=\sum _{i,j}{\int }_u{\int }_v\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}vu& \mathrm{E10}\end{array}$

The surface patch function may be any function that is continuous in its first derivatives. In the embodiment shown, a cubic B-spline interpolating function is used for the interpolating surface patch function although any surface function may be used. This interpolating function is applied to each of the Cartesian coordinate functions shown in equation E5. The interpolating equation for the x-coordinate of the s_i,j patch function is given in equation E11.

x_i,j(u,v)=uM_bX_i,jM_b^tv^t  E11

$u=\left[\begin{array}{c}{u}^3\\ u^2\\ u\\ 1\end{array}\right],v=\left[\begin{array}{c}{v}^3\\ v^2\\ v\\ 1\end{array}\right],$

where t denotes matrix and vector transpose,

$M_b=\left[\begin{array}{cccc}-1& 3& -3& 1\\ 3& -6& 3& 0\\ -3& 0& 3& 0\\ 1& 4& 1& 0\end{array}\right],\mathrm{and}$ $X_{i,j}=\left[\begin{array}{cccc}{x}_{i-1,j-1}& x_{i-1,j}& x_{i-1,j+1}& x_{i-1,j+2}\\ x_{i,j-1}& x_{i,j}& x_{i,j+1}& x_{i,j+2}\\ x_{i+1,j-1}& x_{i+1,j}& x_{i+1,j+1}& x_{i+1,j+2}\\ x_{i+2,j-1}& x_{i+2,j}& x_{i+2,j+1}& x_{i+2,j+2}\end{array}\right]$

Similar calculations are performed for the y_i,j and z_i,j components of the surface patch function.

Since the interpolating functions for each of the patch functions is a cubic surface, the integration may be performed exactly using a quadrature formula. The formula used in this application is shown in equation E12.

$\begin{array}{cc}A\left(s_{i,j}\right)=\sum _{i,j}\frac{1}{4}\left(\begin{array}{c}{\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}}_{u=\frac{3-\sqrt{3}}{6},v=\frac{3-\sqrt{3}}{6}}+\\ {\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}}_{u=\frac{3-\sqrt{3}}{6},v=\frac{3+\sqrt{3}}{6}}+\\ {\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}}_{u=\frac{3+\sqrt{3}}{6},v=\frac{3-\sqrt{3}}{6}}+\\ {\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}}_{u=\frac{3+\sqrt{3}}{6},v=\frac{3+\sqrt{3}}{6}}\end{array}\right)& \mathrm{E12}\end{array}$

Recalling the fact that s_i,j(u,v) is defined as a vector function in Cartesian coordinates (Equation 1), the norm of the cross product of the partial derivatives can be written as follows in equation E13.

$\begin{array}{cc}\frac{\partial s_{i,j}}{\partial u}\times \frac{\partial s_{i,j}}{\partial v}=\sqrt{\begin{array}{c}{\left(\frac{\partial y_{i,j}}{\partial u}\frac{\partial z_{i,j}}{\partial v}-\frac{\partial z_{i,j}}{\partial u}\frac{\partial y_{i,j}}{\partial v}\right)}^2+\\ {\left(\frac{\partial z_{i,j}}{\partial u}\frac{\partial x_{i,j}}{\partial v}-\frac{\partial z_{i,j}}{\partial u}\frac{\partial x_{i,j}}{\partial v}\right)}^2+{\left(\frac{\partial x_{i,j}}{\partial u}\frac{\partial y_{i,j}}{\partial v}-\frac{\partial y_{i,j}}{\partial u}\frac{\partial x_{i,j}}{\partial v}\right)}^2\end{array}}& \mathrm{E13}\end{array}$

When the physical x-, y-, and z-locations are used in the interpolating function, the surface are will be calculated in the square of the units of x, y, and z. At this point the calculation in the fifth block 206-10 of FIG. 20 is complete.

The second component to the mass calculation is a measurement of the thickness of the bladder muscle wall. This thickness is defined to be the normal thickness between the subserosal and submucosal surfaces of the bladder wall.

The wall thickness is calculated from the fractal dimension of the RF signal in the region of the wall thickness. The fractal dimension increases due to the multiplicity of interface reflections through the bladder muscle. The increase and decrease of fractal dimension through the bladder muscle wall can be modeled as a parabola where the fractal dimension is a function of the depth in the region of the bladder wall. The thickness of the bladder is then determined to be the region of the parabola model that is at least 97% of the maximal value of the fractal dimension. The calculations are reviewed below in equation E14.

$\begin{array}{cc}{\mathrm{fd}}_r=\frac{\log \left(\frac{\begin{array}{c}\max \left({\mathrm{RF}}_{r=r-w/2,r+w/2}\right)-\\ \min \left({\mathrm{RF}}_{r=r-w/2,r+w/2}\right)+w\end{array}}{w}\right)}{\log \left(\frac{n}{w}\right)}& \mathrm{E14}\end{array}$

The wall thickness is calculated from the fractal dimension of the RF signal in the region of the wall thickness. The fractal dimension increases due to the multiplicity of interface reflections through the bladder muscle. The increase and decrease of fractal dimension through the bladder muscle wall can be modeled as a parabola where the fractal dimension is a function of the depth in the region of the bladder wall. The thickness of the bladder is then determined to be the region of the parabola model that is at least 97% of the maximal value of the fractal dimension. The calculations are reviewed below in equation 15.

The fractal dimension calculation corresponds to the fourth block 206-12 of FIG. 20. The fractal dimension is calculated for a window of length w. In the current embodiment the value of w is 5, the number of sample points along a scan line, although that value can be varied. The fractal dimension is calculated from the difference between the maximum RF signal value in the window centered at a given depth, r, and the minimum of that same window. The length of the window, w, is added to this difference, and the result is then normalized with the length of the window. The logarithm of that result is then divided by the logarithm of the ratio of the total number of samples in a scan line, n, to the length of the window. The calculation of the fractal dimension at each depth along a scan line is shown in Equation 10. This fractal dimension measure is calculated for the central n-w samples in a scan line.

After the measurements of the fractal dimension have been calculated based on the ultrasound signal, the thickness of the bladder wall may be calculated. The following calculations correspond to the seventh block 206-16 of FIG. 19.

The fractal dimension, fd, of the RF signal in the region of the bladder muscle wall is then modeled as a parabolic equation as a function of depth, r. The model of the equation for a single depth point is given in equation E15. In that equation, there are 3 parameters (a, b, and c) that define the parabola with the depth along a scan line r, and the addition of a random element ε. The subscript i indicates a specific value of r,fd, and ε.

fd_i=ar_i²+br_i+C+ε_i  E15

An equation of the form in equation E15 is obtained for each depth point in the region of the wall. The number of observations is variable and depends on the thickness of the bladder wall as observed by the ultrasound signal. Assuming a set of n observations, the subscript i would count the observations from 1 to n. The set of n equations of the form in equation 15 may be compressed into a matrix equation given in equation E16.

$\begin{array}{cc}\mathrm{fd}=X\beta +\varepsilon \mathrm{where}\mathrm{fd}=\left[\begin{array}{c}{\mathrm{fd}}_1\\ {\mathrm{fd}}_2\\ ⋮\\ {\mathrm{fd}}_n\end{array}\right],X=\left[\begin{array}{ccc}{r}_1^2& r_1& 1\\ r_2^2& r_2& 1\\ ⋮& ⋮& ⋮\\ r_n^2& r_n& 1\end{array}\right],\beta =\left[\begin{array}{c}a\\ b\\ c\end{array}\right],\mathrm{and}\varepsilon =\left[\begin{array}{c}{\varepsilon }_1\\ {\varepsilon }_2\\ ⋮\\ {\varepsilon }_n\end{array}\right]& \mathrm{E16}\end{array}$

Each row of the fd, and ε, and the X matrix correspond to one of the n observations. The parabola parameters of equation E16 are collected in the vector P.

The next step is to estimate the values of the parameters of the parabola in the set of n equations of the form in equation E15 or in the matrix equation E16 based on the set of observations. A least-squares estimation of the parameters is used, and the calculation for these estimates is shown in equation E17. In E17, the t superscript indicates matrix transpose, and the −1 superscript indicates the matrix inverse. Parameters with hats (A) indicate that the value is the least-squares estimate of those parameters.

{circumflex over (β)}=(X^tX)⁻¹X^tfd  E17

The estimates of the parabola parameters ({circumflex over (β)}=└â {circumflex over (b)} ĉ┘′) can be substituted into the parabola model to calculate the estimated fractal dimension at each depth r, as shown in equation E18. The location of the maximum fractal dimension can be determined by setting the first derivative of the parabola model to equal 0 (equation E19) and solving for r. The location where the fractal dimension is maximal is given in equation E20.

$\begin{array}{cc}f\hat{d}\left(r\right)=\hat{a}r^2+\hat{b}r+\hat{c}& \mathrm{E18}\\ \frac{f\hat{d}\left(r\right)}{r}=2\hat{a}r+\hat{b}=0& \mathrm{E19}\\ r_{{\mathrm{fd}}_{\max }}=-\frac{\hat{b}}{2\hat{a}}& \mathrm{E20}\end{array}$

To determine the maximal fractal dimension as defined by the parabolic model, simply substitute equation 20 into equation 18 and solve for fd_max. The resulting value is shown in equation E21.

$\begin{array}{cc}f{\hat{d}}_{\max }=\frac{-{\hat{b}}^2+4\hat{c}}{4\hat{a}}.& \mathrm{E21}\end{array}$

To determine the locations where the fractal dimension is 97% of the maximum value, multiply equation E21 by 0.97, substitute the result into equation E18 and solve for r using the quadratic formula. The locations where the fractal dimension is 97% of the maximum value, r_97%, are given in equation E22.

$\begin{array}{cc}{r}_{97\%}=\frac{-\hat{b}\pm \sqrt{{\hat{b}}^2-4\hat{a}\left(\hat{c}+0.97\frac{{\hat{b}}^2+4\hat{c}}{4\hat{a}}\right)}}{2\hat{a}}& \mathrm{E22}\end{array}$

Two values for r_97% will be calculated from Equation 18. The difference between those two values will identify the thickness of the bladder muscle wall along the given scan line. Since these scan lines may or may not be perpendicular to the bladder muscle surface and bladder wall thickness must be measured along a line perpendicular to the bladder surface, a collection of these measurements are combined to determine the actual thickness of the bladder wall.

These measurements could be made at any surface of the bladder muscle wall. In FIG. 22, three scan lines are shown to cross the bladder muscle in two locations: the anterior wall closest to the transducer, and the posterior wall furthest from the transducer. The parabolic model described previously can be applied twice on each to determine the thickness of both the anterior and posterior wall. The maximum and minimum and mean values of these thicknesses are used in the mass calculation and historical tracking of data. In the embodiment shown, this final thickness determination marks the end of the process identified in the seventh block 206-16 of FIG. 19.

FIG. 21 is a schematic representation of three scan lines passing through the subserosal and submucosal wall locations of an organ, here schematically illustrated for a bladder. Three scan lines 362, 364, and 366 penetrate the bladder. The dotted portion of the lines represents the portion of the scan lines that passes through the bladder muscle wall at an anterior or front wall location 370A and a posterior or back wall location 370B. The first 362, the second 364, and the third 366 scan lines are shown transmitting through the front subserosal wall location 372A and front submucosal wall location 374A. Similarly, the first 362, the second 364, and the third 366 scan lines are shown transmitting across the internal bladder region 375 and through the back submucosal wall location 374B and back submserosal wall location 372B. The front and back subserosal locations 372A and 372B occupy an outer bladder wall perimeter and the front and back submucosal locations 374A and 374B occupy an inner bladder wall perimeter. A bladder wall thickness value 376 is obtained for the respective differences along each scan line 362-366 between the subsersosal wall locations 372A and the submucosal wall locations 374A, or the subserosal wall locations 372B and the submucosal wall locations 374B. The maximum and minimum and mean values of these thicknesses are used in the bladder wall mass calculation and historical tracking of data. In the preferred embodiment, the bladder is assumed to have a uniform wall thickness, so that a mean wall thickness value is derived from the scanned data and used for the determination of the internal wall volume 375. Only three scan lines are shown in a plane, each separated by 7.5 degrees from each other. Both the number of scan lines in the plane and the angles separating each scan line within a plane may be varied.

Once the bladder wall thickness and the inner and outer surface area have been measured, the volume of the internal bladder region 375 may be calculated by the determining the respective differences between the front and back submucosal wall locations 374A and 374B along each scanline penetrating the bladder region 375. The difference between the front and back submucosal wall locations 374A and 374B defines an inter-submucosal distance. The internal volume of the bladder region 375 is then calculated as a function of the inter-submucosal distances of the penetrating scan lines and the area of the subserosal boundary or internal bladder perimeter. The volume of internal region 375 is assumed to be the surface area times a function of the inter-submucosal distances, where the assumption is further based on a uniform wall subserosal boundary at all points around the internal bladder perimeter. In the embodiment shown, this volume calculation corresponds to the eighth block 206-20 of FIG. 19.

The methods to obtain the wall-thickness data, the mass data, and the volume of internal region 375 via downloaded digital signals can be configured by the microprocessor system for remote operation via the Internet web-based system. The Internet web-based system (“System For Remote Evaluation Of Ultrasound Information Obtained By A Program Application-Specific Data Collection Device”) is described in patent application Ser. No. 09/620,766, herein incorporated by reference. The internet web-based system has multiple programs that collect, analyze, and store organ thickness and organ mass determinations. The alternate embodiment thus provides an ability to measure the rate at which internal organs undergo hypertrophy with time and permits disease tracking, disease progression, and provides educational instructions to patients.

In summary, the foregoing method supplements current algorithms in novel ways that advantageously allow different bodily fluids and/or tissues to be distinguished by volume and composition. In particular, different regions having low echogenicity may be properly distinguished. This feature advantageously permits shadowed regions of a fetus such as the arms and the legs of the fetus to be distinguished from a head region. In a diagnostic method directed to the detection of an aortic aneurysm, the foregoing embodiments may be used to differentiate shadowed regions resulting from bowel gas from other low echo regions. In a gall bladder imaging scan, the foregoing embodiments may be used to determine whether bile or other bodily fluids are within the field of view of the ultrasound-scanning device.**

The first method describes a simple device that allows the assessment of bladder 23 volume, using only a few beams appropriately oriented. Under the assumption that there exists a correlation between the bladder height and width, a simple approach has been developed. It consists of a limited number of acoustic beams positioned in such a way that the depth D and the height H of the bladder could be estimated in approximately a single sagittal plane. The volume of urine is then computed from an empirical formula that does not assume any geometric model.

In operation of the apparatus of the present invention, the transducer assembly 1 is placed on the abdomen of the patient in the supine position, just above the symphysis pubis 2. We are presenting a particular configuration of the assembly 1. Nevertheless, various configurations can be derived from this model and several modifications could be achieved (number of transducers, position, orientation, etc. . . . ) without departing from the initial ideas. The device proposed as an example is composed of five disc shaped transducers A, B, C, D and E (focused or non-focused) positioned in the assembly at predetermined distance from each other (FIG. 5, top panel) and oriented at predetermined angles φ_A, φ_B, φ_C, φ_D, and φ_E (FIG. 6). Referring to FIG. 5 (top panel), it appears that the transducers A, B, C, D and E are oriented in two different planes. The distance between these two planes is small compared to the bladder 3 size and thus we can assume that the information received from each transducer represent the characteristics of approximately a single sagittal or anteroposterior plane. The orientation of each beam has been determined from the knowledge of the bladder 3 position and shape when it is filling up as measured in a patient series. The first beam of the transducer assembly 1 (soundbeam from transducer A) is oriented in such a way that it reaches the bottom of the bladder, passing just above the symphysis pubis 2. The remaining beams are positioned for successively intercepting the bladder 3 when it expands with increasing filling degree.

Computation of the Depth D and Height 5: Depending on the number of beams that are intercepting the bladder 3 and on the geometrical configuration of the transducer assembly (1), the distances H and D are determined by different mathematical procedures. For most measurement configurations the depth D of the bladder is determined by the distance between echoes derived from front and back wall of the bladder estimated from Transducer B.

The Height H (5) calculation in the specific measurement configuration (here we selected as an example the “filled bladder” configuration of FIG. 2) when posterior bladder wall echoes are detected in signals obtained in beam A, B, C, and D, but not in beam E is illustrated in FIG. 3. For the other filling geometries the height is calculated in a corresponding way. The mathematical procedure is as follows:

$\begin{array}{cc}\cos {\varnothing }_A=\left[\mathrm{AA}2\right]/\left[\mathrm{AA}1\right]\text{=>}\left[\mathrm{AA}2\right]=\cos {\varnothing }_A·\left[\mathrm{AA}1\right]& \left(1\right)\\ \sin {\varnothing }_A=\left[A1A2\right]/\left[\mathrm{AA}1\right]\text{=>}\left[A1A2\right]=\sin {\varnothing }_A·\left[\mathrm{AA}1\right]& \left(2\right)\\ \cos {\varnothing }_D=\left[\mathrm{DD}2\right]/\left[\mathrm{DD}1\right]\text{=>}\left[\mathrm{DD}2\right]=\cos {\varnothing }_D·\left[\mathrm{DD}1\right]& \left(3\right)\\ \cos {\varnothing }_A=\left[D1D2\right]/\left[\mathrm{DD}1\right]\text{=>}\left[D1D2\right]=\sin {\varnothing }_D·\left[\mathrm{DD}1\right]& \left(4\right)\\ \cos {\varnothing }_A··=\left[\mathrm{AA}2\right]/\left[\mathrm{AA}1\right]\text{=>}\left[\mathrm{AA}2\right]=\cos {\varnothing }_A·\left[\mathrm{AA}1\right]& \left(5\right)\\ \cos {\varnothing }_A··=\left[\mathrm{AA}2\right]/\left[\mathrm{AA}1\right]\text{=>}\left[\mathrm{AA}2\right]=\cos {\varnothing }_A·\left[\mathrm{AA}1\right]& \left(5\right)\\ \mathrm{ID}1=\left[D1D2\right]+\left[A1A2\right]+\left[\mathrm{AD}\right]& \left(6\right)\\ \text{=>}\mathrm{Height}=\left[A1D1\right]=\sqrt{{\left[A1I\right]}^2+{\left[\mathrm{ID}1\right]}^2}& \left(7\right)\end{array}$

Volume computation: The volume of urine is correlated to the bladder diameter (Height 27 and Depth 26) by the empirical formulae:

Height*Depth*K

where K is a correction factor. Depending on the number of beams that allow the determination of the bladder dimensions (from 1 to 5) and others parameters such as the age, the gender, the correction factor is different. For a given situation (parameters other than number of beam are fixed), the correction factors KL, K2, K3, K4 and K5 are optimized using linear regression analysis.

The process executed by the hardware is illustrated in the flow chart of FIG. 4.

After positioning the transducer assembly correctly over the bladder area the measurement procedure is started by pressing the start button which during the (short) measurement procedure remains depressed. Subsequently the transducers are activated for transmission of ultrasound pulses and reception of echoes and possible detection of bladder wall echoes in a specific order. Thereafter it is established, when a clear posterior bladder wall echo is detected, which ultrasound beams, this we call here the beams of “useful” transducers, penetrate the filled bladder. From this, the filling situation or measurement geometry is established. As a result the proper correction factor can be selected. After calculation of the volume the value is stored in memory and displayed. During the measurement procedure the transducer assembly is slightly moved and memory data are refreshed if a larger volume is measured. The highest value will correspond with the correct bladder volume. This is displayed.

In a general aspect, therefore, the apparatus may use beam information comprising at least: angle of incidence (known from the transducer mounting angle), spatial position (known from the transducer position in the array) and echo travel time (deduced from the reflected beam). Other beam parameters or information from reflected beams may also be used in accordance with known ultrasound techniques, such as frequency, pulse rate etc.

For determining body cavity and height, the apparatus may select only beams corresponding to those that have intercepted the fluid filled body cavity.

The arrangements described in connection with FIGS. 1 to 6 illustrate use of five transducers. This configuration was selected in order to achieve a selected degree of accuracy of measurement over a complete expected range of total volumes in a human adult. In the preferred configuration, accuracy of measurement of the order of 100 ml over a range encompassing a bladder fill level from 0 to approximately 800 ml has been exhibited. It will be understood that a smaller number of transducers could be used when either the desired measurement accuracy can be reduced, or when the total fill range covered can be reduced.

For example, using just three transducers, it has been shown to be possible to cover a fill range of 0 to approximately 500 ml with an accuracy of 100 ml.

Similarly, four transducers has been shown to cover a range 0 to approximately 700 ml, and two transducers, a range of 0 to approximately 300 ml.

Such configurations can be used when it is only necessary to indicate gross ranges of bladder filling, or to indicate a clinically important threshold fill level.

In other embodiments, the apparatus may be provided with an input device such as a keypad or computer interface so that the user can enter patient information, such as gender, weight and age. This information can then be used to ensure correct selection of an available correction factor, K, from a memory of the apparatus.

The apparatus may also be provided with means for inputting calibration data, such as absolute measurements of bladder fill level separately deduced from conventional measurements. These can be stored by the apparatus and used to optimise stored K values as part of an iterative, ‘self-learning’ process. In other words, the apparatus may incorporate an algorithm for automatically adjusting predetermined correction factors stored therein based on calibration data entered into the machine for comparison with measurement data taken by the apparatus.

The apparatus may also comprise a means for indicating correct caudal-cranial positioning of the transducer array on the body over the bladder. For example, in a normal measurement as suggested in figure, it is expected that at least transducers A, B and C will indicate a bladder present condition, whereas transducers D and E might, or might not indicate bladder present, according to the bladder fill level. In the event that, for example, no signal is indicated by A, or by A and B, but signal is indicated by D or D and E, then it can be deduced that the transducer assembly is positioned too far in the cranial direction. This could be indicated on the display of the device.

In summary, the described first method differs greatly from known other apparatus:

1) The device is composed of a limited number of static single element transducers;

2) The arrangement of the transducer is not similar to the arrangement of a linear array;

3) The transducers are oriented towards the bladder with specific angles allowing the estimation of the urine volume over a wide range of volumes;

4) The method for automatic volume computation does not assume any geometrical model for the bladder shape;

5) It is valid for any bladder shape since the volume is computed with an empirical formula for various filling ranges;

6) It is not based only on the measurement of distances between the front and back wall or area in different planes;

7) It uses an automatic detection of the bladder height and depth depending on the number of beams that intercept the bladder;

8) It optimizes the correction factor depending on the degree of filling (or other factors, such as age, gender, weight, that may influence the calculations);

9) The device includes a closed loop to easily find the optimal position;

10) The optimal position corresponds to the largest volume computed;

11) The device works instantaneously.

DETAILED DESCRIPTION OF THE SECOND METHOD

The second version of the device is based on a different principle. The approach consists of using a single acoustic beam with a very wide width such that it encloses approximately the entire volume of the bladder when it is filled up. Such a wide beam width can be obtained using a single element transducer with a defocusing lens as drawn in FIG. 7 or a curved single element transducer.

The schematic principle of transducer positioning is illustrated in FIG. 7. The sagittal cross section through the bladder is shown. The cone like shape of the acoustic beam allows to encompass approximately the full bladder volume, and therefore any harmonic distortion detected in the echo signal returning from a region beyond the posterior wall of the bladder around depth W, would correlate to the amount of fluid contained in the bladder.

It has been demonstrated that the propagation of ultrasound waves is a nonlinear process. The nonlinear effects, which increase with higher intensities, have been predicted and demonstrated at frequencies and intensities used in the diagnostic range either in water or in human body (A Baker et al.: “Distortion and High-Frequency Generation Due to Non-Linear Propagation of Short Ultrasonic Pulses From A Plane Circular Piston”, J. Acoustic Soc Am 92(3), pp 1699-1705). The distortion is due to slight non-linearities in sound propagation that gradually deform the shape of the propagating sound wave, and result in development of harmonic frequencies which were not present in the transmitted wave close to the transducer. This manifests itself in the frequency domain as the appearance of additional harmonic signals at integer multiples of the original frequency.

These effects occur most strongly when ultrasound waves propagate within liquids with relatively low acoustic attenuation such as water, amniotic fluid or urine. Indeed, acoustic propagation in fluids gives rise to extreme nonlinear effects at diagnostic frequencies. Within soft tissues, nonlinear processes also take place but are modified as a result of the different acoustic characteristics of these tissues, most notably their high acoustic absorption. Indeed, water and amniotic fluids (urine) are significantly different from tissue.

It is known from literature (A C Baker: “Prediction Of Non-Linear Propagation In Water Due To Diagnostic Medical Ultrasound Equipment”, Phys Med Biol 1991 VOL 36, NO 11, PP 1457-1464; T Szabo et al.: “Effects of Non-Linearity On The Estimation Of In-Situ Values Of Acoustic Output Parameters”, J Ultrasound Med 18:33-41,1999; M Hamilton et al.: “Nonlinear Acoustics”, Academic Press) that the non-linearity of a medium is characterized by the coefficient of non-linearity B. Typical values for P are 3. 6 for water, 4 for blood and 6.5 for fatty tissue.

In addition to being nonlinear, all the media have acoustical loss due to absorption. The acoustical loss is described by the power law: A=AOFB where ao is constant and b ranges from 1 to 2 depending on the medium. For water, the rate of absorption of an ultrasound wave propagating through it is quadratically related to the frequency (b=2). However, the rate of energy loss due to absorption is considered small and most of the time the dissipation-less theory is applicable over short ranges. However, biological media have large rates of energy loss and the frequency dependence has an exponential value of 1 to 1.5.

By considering both attenuation due to absorption loss and non-linearity, the exchange of energy between the two processes is complicated, because attenuation diminishes the amplitude of the generated harmonic components with propagation distance while non-linearity builds up these harmonics. So, harmonic distortion generally tends to enrich the higher harmonic components at the expense of the lower ones (energy transfer), while absorption damps out the higher components more rapidly than the lower ones. It is therefore difficult to reach a balance in which a given component loses as much energy by absorption as it gains from nonlinear distortion. Moreover, since the conditions for stability depend on the amplitude of the wave, which slowly decreases with propagation distance, the wave can never be completely stable, only relatively so.

The balance between the nonlinear process and the attenuation process is given by the Goldberg number Γ (Szabo et al.), which represents a measure of which process dominates. When Γ=1, nonlinear effects are comparable to attenuation effects. If Γ is higher than 1, nonlinear processes dominate and when the Goldberg number is below 1, attenuation effects take over. As indication, for acoustic pressures of 500 kPa and LMPA, at a transmit frequency of 3 MHz, the Goldberg number is respectively 86.5 and 43.2 for water. It is only 2.8 and 1.4 for liver-like tissue respectively at these pressures. For both settings, the parameter shows that for water, non-linearity is up to thirty times greater than for tissue.

The approach used here is based on the “non-linearity/attenuation” characteristic in differentiating between fluid media and soft tissue media. As described above, a single element transducer is placed in front of the bladder. The transducer generates a wide acoustic beam that is able to enclose the full bladder volume. Depending on the volume of urine contained in the bladder (bladder filling) and thus crossed by the acoustic beam, the amount of harmonic distortion generated in the back of the bladder will change. A radio frequency (RF) backscattered signal might be selected from a region of interest located preferably in the backside of the bladder. The amount of energy comprised in the second harmonic or higher harmonic components of the received RF echo signal can be extracted and correlated to the amount of volume of urine that has been encompassed by the acoustic beam. Since harmonic generation is different in tissue than in fluids, only the volume of urine that has been crossed by the acoustic beam would generate more harmonic energy. When the bladder is empty or below a certain volume level, no harmonic distortion occurs, whereas maximal distortion will be obtained for a full volume.

FIG. 9 illustrates the principle of the invention. Top panel shows two situations. The bladder is either empty (Panel A left side) or filled up with urine (Panel A right side). At a certain distance beyond the bladder (around 12 cm from the transducer), a region of interest of 1.5 cm width at depth W (see FIG. 7) is selected. Power spectra corresponding to echo signal recorded from the regions of interest are displayed in panel B.

The spectrum corresponding to the empty bladder (solid line) shows only a fundamental component. The harmonic distortion is very weak so that no harmonic frequencies are generated. However, the echo signal corresponding to the filled bladder situation (dashed line) demonstrates clear distortion where a second harmonic component with a significant energy is generated. The third harmonic component can be also present with lesser energy depending on the urine volume that has been crossed by the acoustic beam.

FIG. 9 demonstrates that depending on the volume contained in the bladder that the acoustic beam has intersected, the amount of generated second harmonic energy varies. When the acoustic beam crosses only tissue or when the volume of urine is very small, harmonic distortion is the lowest with no or very low harmonic energy. If the bladder is filled up or if the volume of urine is above a certain level (threshold), harmonics are generated. The generation of a harmonic component (second and/or higher harmonics) can be used for volume measurement, or simply as an indicator of filling of the bladder to a certain volume extent. The criterion can be such that if a certain amount of second harmonic (or higher harmonics) is generated in the echo signal, the device would indicate that the critical volume (or threshold) (say in adult patients around 450 ml) has been reached.

To avoid and eliminate any differences due to patient to patient variations, a normalization procedure needs to be performed a priori. Such a normalization procedure might consist of recording a first signal at very low transmit acoustic power from the same region of interest as described in the previous section. Such power would allow only linear propagation of the ultrasonic waves and avoid any harmonic generation. The echo signal would therefore have undergone only attenuation effects.

In the following transmit-receive sequence, the transmit acoustic power is increased with a certain factor (e) and a new recording is performed from the same region of interest. This measure with a much higher acoustic pressure is carried out to allow harmonic distortion to occur in the tissue. The echo signal in this case will undergo both attenuation and distortion effects. The first echo signal (linear case) will be re-scaled by the factor that corresponded to the increase in transmit power (e), and then used as a reference signal. Consequently, each patient has his own reference hence eliminating any variations such as obesity, INHOMOGENEITIES, etc.

A block diagram of a possible steps describing the second method is given in the flow chart of FIG. 8. The two transmitted signals might be transmitted with a very low repetition rate as indicated in FIG. 10. The first packet of transmit signals with low acoustic amplitude are used for calibration. The echoes received from those signals are averaged to reduce the noise level.

The number of signals can be chosen such that a high signal-to-noise ratio is obtained. The second packet of signals with higher amplitudes are used to induce nonlinear propagation and harmonic distortion. The echoes received from these signals are averaged and then the harmonic energy is filtered and then compared to the calibration echo.

In order to estimate the volume of urine in the bladder, a look-up table can be created beforehand. Such a table, saved in the hard disk of the electronic device, will contain the correspondence between the harmonic energy and the volume of urine. Such a table can be extracted from a curve similar to the one given in FIG. 11. Such a curve can be obtained from a “learning” patient set of measurements. Look-up tables may eventually be produced for specific patient groups for age; gender and/or weight as an input parameter.

The described second method differs greatly from known other apparatus:

12) The device is composed of a single element defocused ultrasound transducer with a conical beam shape;

13) The single acoustic beam entirely encompasses the volumetric area of a possibly filled bladder.

14) The method is based on measurement of non-linear properties and attenuation behavior of propagating ultrasound waves as influenced by a urine filled bladder.

15) The method incorporates a technique to eliminate patient variation due to fat or skin properties.

16) The method for automatic volume computation does not assume any geometrical model for the bladder shape;

17) It is valid for any bladder shape since the received signal “integrates” all volume effects in the ultrasound beam.

18) All known other methods use bladder wall echoes as a basis to calculate volume.

19) The device works instantaneously. Other embodiments are intentionally within the scope of the accompanying claims.

While preferred and alternate embodiments of the invention have been illustrated and described, as noted above, many changes can be made without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is not limited by the disclosure of these preferred and alternate embodiments. Instead, the invention should be determined entirely by reference to the claims that follow.

Claims

1. An apparatus for measuring the volume of fluid in a human or animal body cavity using a non-invasive, ultrasound echo technique, comprising: a transducer assembly including a plurality of ultrasound transducers mounted thereon for transmitting and receiving a plurality of ultrasound signals into the body cavity at plural angles of incidence and/or from plural spatial locations; means for activating the transducers to produce transmitted ultrasound signals; means for detecting body cavity wall echoes from received ultrasound signals; means for determining, from said received signals, a body cavity height H and depth D; means for determining a specific measurement configuration corresponding to the body cavity filling degree from the ultrasound signals that intercept the fluid filled body cavity to thereby select an appropriate predetermined correction factor K corresponding to that specific measurement configuration, for optimal calculation of the volume; and means for calculating the fluid volume according to the formula H×D×K.

2. The apparatus of claim 1, wherein the body cavity is a bladder and the volume of fluid measured is a volume of urine.

3. The apparatus of claim 1, wherein the means for activating includes means for transmitting said plurality of ultrasound signals in a selected order.

4. The apparatus of claim 1, wherein the means for detecting uses echo travel time and other beam information from the plurality of ultrasound signals.

5. The apparatus of claim 1, wherein the means for determining selects specific ultrasound signals from the plurality of ultrasound signals corresponding to ultrasound beams that have intercepted the fluid filled body cavity.

6. The apparatus of claim 1, further including a display means for instantaneous display of the calculated fluid volume to allow optimization of transducer positioning by the user.

7. The apparatus of claim 1, wherein the means for deriving includes a memory storing a plurality of empirically predetermined correction factors K.

8. The apparatus of claim 1, wherein the array further includes five transducers.

9. The apparatus of claim 8, wherein the five transducers are respectively oriented at angles φA, φB, φC, φD, and φE, to an axis orthogonal to the plane of the transducer array, the angles being approximately φA=−25°, φB=0°, φC+25°, φD+25°, φE+40°.

10. A method for measuring the volume of fluid in a human or animal body cavity using a non-invasive, ultrasound echo technique, comprising the steps of: transmitting a plurality of ultrasonic beams into the region of the body containing the cavity at plural angles of incidence and/or from plural spatial locations; receiving a plurality of ultrasonic signals from the body; determining, from said received signals, a body cavity height H and depth D; determining, from the received signals, a specific measurement configuration corresponding to the body cavity filling degree from the ultrasound signals that intercept the fluid filled body cavity to thereby select an appropriate predetermined correction factor K corresponding to that specific measurement configuration, for optimal calculation of the volume; and calculating the fluid volume according to the formula H×D×K.

11. The method of claim 10, further including the step of transmitting the plurality of ultrasonic beams into the body from a transducer array in which a plurality of transducers are arranged with a predetermined spatial location and mounting angle.

12. An ultrasonography method, comprising:

creating a database that is representative of a tissue, a fluid, or a cavity of a body;

transmitting ultrasound pulses into a region-of-interest in a patient;

receiving echoes from the region of interest, and based upon the received echoes: compiling an ultrasonic pattern of the region-of-interest; processing the pattern by comparing the region-of-interest patterns to the database; and

determining a composition within the region-of-interest of the patient.

13. The method of claim 12, wherein transmitting ultrasound pulses into the region of interest includes transmitting the pulses to at least one of a tissue, a fluid, and a cavity.

14. The method of claim 13, wherein transmitting the pulse further comprises transmitting the pulses to at least one of urine, blood, amniotic fluid, lung fluids, liver bile, and mixtures thereof.

15. The method of claim 12, wherein processing the pattern includes calculating at least one of a Goldberg number, a harmonic ratio, and an attenuation factor.

16. The method of claim 15, wherein processing the pattern further includes applying a window algorithm to a section of an echo pulse near a cavity-boundary interface within the region-of-interest.

17. The method of claim 16, wherein applying a window algorithm further comprises determining the harmonic frequencies associated with the section of the echo pulse near the cavity-boundary interface.

18. The method of claim 12, wherein receiving echoes further comprises receiving at least one of a single dimensional line, a two-dimensional plane, and a three-dimensional array of two-dimensional planes.

19. An ultrasonography method, comprising:

creating a database that is representative of a tissue, a fluid, and a cavity of a body;

transmitting ultrasound pulses into a region-of-interest in the body;

receiving echoes from the region of interest, and based on the echoes: compiling an ultrasonic pattern of the region-of-interest; processing the pattern by comparing the region-of-interest patterns to the database; and

determining a volume within the region-of-interest of the body.

20. The method of claim 19, wherein transmitting ultrasound pulses to the region of interest includes transmitting the pulses into at least one of a tissue, a fluid, and a cavity.

21. The method of claim 20, wherein transmitting the pulse further comprises transmitting the pulses into at least one of urine, blood, amniotic fluid, lung fluids, liver bile, and mixtures thereof.

22. The method of claim 19, wherein processing the pattern includes calculating at least one of a Goldberg number, a harmonic ratio, and an attenuation factor.

23. The method of claim 19, wherein processing the pattern further includes applying a window algorithm to a section of an echo pulse near a cavity-boundary interface within the region-of-interest.

24. The method of claim 23, wherein applying a window algorithm further comprises determining one or more harmonic frequencies associated with the section of the echo pulse near the cavity-boundary interface.

25. The method of claim 19, wherein receiving echoes further comprises receiving at least one of a single dimensional line, a two-dimensional plane, and a three-dimensional array of two-dimensional planes