SYSTEMS, PROCESSES AND SOFTWARE ARRANGEMENTS FOR EVALUATING INFORMATION ASSOCIATED WITH AN ANATOMICAL STRUCTURE BY AN OPTICAL COHERENCE RANGING TECHNIQUE

Abstract

Software systems, arrangements and processes for evaluating an image associated with at least one portion of an anatomical structure are provided. For example, first information associated with the at least one portion of the anatomical structure second information associated with the at least one portion of the anatomical structure can be received. Third information can be generated by determining a relationship between the first information and the second information. Further, the image can be evaluated using a predetermined pathological scoring criteria and the third information.

Description

CROSS-REFERENCE TO RELATED APPLICATION(S)

This application is a continuation application of U.S. patent application Ser. No. 13/735,580 filed on Jan. 7, 2013 which will issue as U.S. Pat. No. 9,326,682 on May 3, 2016, which claims priority from U.S. patent application Ser. No. 11/414,564, filed Apr. 28, 2006 which issued as U.S. Pat. No. 8,353,665 on Jan. 8, 2013, and claims priority from U.S. Patent Application Ser. No. 60/676,362, filed Apr. 28, 2005, the entire disclosures of which are incorporated herein by reference.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

The invention was made with the U.S. Government support under Contract No. RO1 CA 103767 awarded by the National Institute of Health. Thus, the U.S. Government has certain rights in the invention.

FIELD OF THE INVENTION

The present invention relates to systems, processes and software arrangements for evaluating information associated with an anatomical structure by optical coherence ranging evaluating information by an optical coherence ranging technique, and for example for interpreting microscopic images obtained from living subjects.

BACKGROUND INFORMATION

A variety of different optical biopsy techniques have been described and developed for a non-invasive diagnosis of disease within living human patients. While these conventional devices can provide information that is related to disease, there are differences between the data obtained by these prior art methods and the medical standards of care for diagnosis.

Pathologists generally diagnose tissues based on a microscopic visualization of Hematoxylin & Eosin (H&E) stained slides and a morphological interpretation thereof. Pathologists may employ scoring systems or techniques, in which a variety of features are noted and formed to render a diagnosis. These scoring systems or techniques can standardize and provide a quantitative or semi-quantitative basis for diagnosis. Examples of such scoring systems and techniques include a Gleason grade for prostate adenocarcinoma, Haggitt's criteria for dysplasia in Barrett's esophagus, Banff kidney allograft scoring system, Nash scoring system for nonalcoholic fatty liver disease. Other scoring systems and techniques for such diagnosis exist.

Preferably, a unique relationship can be established between the optical biopsy information and the techniques and scoring systems that for the basis of the standard of care. In turn, the same criteria are generally used to render the diagnosis for the standard of care can then be utilized, in a modified form, on the optical biopsy diagnostic information. In turn, a modified scoring system or technique based on features identified in the optical biopsy images may be implemented to diagnose tissue in a manner consistent with the histopathology standard of care.

Provided below are examples of upper gastrointestinal scenarios which may be pertinent to Barrett's esophagus, where optical biopsy images can be utilized to render a diagnosis.

Diagnosing Specialized Metaplasia at the Gastroesophageal Junction

Gastroesophageal reflux disease (GERD) is increasing in incidence, and is a well-known risk factor for the development of esophageal specialized intestinal metaplasia (SIM), commonly known as Barrett's esophagus (BE), as described in R. J. Loffeld et al. “Rising incidence of reflux oesophagitis in patients undergoing upper gastrointestinal endoscopy” Digestion, 2003, Vol. 68(2-3) pp. 141-4. The prevalence of SIM has been estimated to be as high as 10-15% in patients with chronic GERD as discussed in C. Winters, Jr. et al., “Barrett's esophagus. A prevalent, occult complication of gastroesophageal reflux disease. Gastroenterology,” 1987, Vol. 92(1), pp. 118-24. For a patient with recurrent and severe symptoms of GERD, the adjusted odds ratio for developing adenocarcinoma over a 20-year period is 7.7 and 43.5, respectively, as described in J. Lagergren et al., “Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma,” N Engl J Med, 1999, Vol. 340(11), pp. 825-31. Moreover, the incidence of esophageal adenocarcinoma and proximal stomach (gastric cardia) cancers has rapidly increased in the last 30 years, as discussed in W. J. Blot et al., “Rising incidence of adenocarcinoma of the esophagus and gastric cardia,” Jama, 1991, Vol. 265(10), pp. 1287-9; P. Bytzer et al., “Adenocarcinoma of the esophagus and Barrett's esophagus: a population-based study,” Am J Gastroenterol, 1999, Vol. 94(1), pp. 86-91; and S. S. Devesa et al., “Changing patterns in the incidence of esophageal and gastric carcinoma in the United States,” Cancer, 1998, Vol. 83(10), pp. 2049-53.

Due to the recognition of GERD as possible a risk factor for developing esophageal cancer, upper endoscopic screening can be recommended, e.g., for white, male patients older than 50 years who have chronic symptoms of GERD for more than 5 years, as discussed in S. J. Spechler, “Screening and surveillance for complications related to gastroesophageal reflux disease,” Am J Med, 2001, Vol. 111 Suppl 8A, pp. 130S-136S. As a result of the increasing prevalence of GERD and the medical community's recognition of SIM as a risk factor for esophageal cancer, the use of endoscopy as a screening strategy for SIM will likely increase significantly in the near future. Such increases may incur significant costs to the health care system and to the individual patient. Other screening methods that could provide greater area coverage than conventional biopsy may reduce the risk and inconvenience of multiple endoscopic procedures. Furthermore, certain methods which do not utilize endoscopy can potentially be conducted at a lower cost, partially alleviating the financial burden of comprehensive screening on the health care system.

Identifying Dysplasia in Patients with Barrett's Esophagus

When BE is diagnosed, a periodic endoscopic surveillance to detect HGD may be recommended. These recommendations can proceed from observations noting the high incidence (25% over 46 months) of adenocarcinoma in patients with HGD, as described in P. Sharma et al., “A critical review of the diagnosis and management of Barrett's esophagus: the AGA Chicago Workshop,” Gastroenterology, 2004, Vol. 127(1), pp. 310-30. Current guidelines for surveillance of HGD can include four-quadrant biopsies every two centimeters along the axial length of the Barrett's segment, as discussed in D. S. Levine et al., “An endoscopic biopsy protocol can differentiate high-grade dysplasia from early adenocarcinoma in Barrett's esophagus,” Gastroenterology, 1993, Vol. 105(1), pp. 40-50. The accuracy of surveillance endoscopy, however, may be limited by a sampling error, as discussed in G. S. Dulai, “Surveying the case for surveillance,” Gastroenterology, 2002, Vol. 122(3), pp. 820-823; G. W. Falk et al., “Surveillance of patients with Barrett's esophagus for dysplasia and cancer with balloon cytology,” Gastroenterology, 1997, Vol. 112(6), pp. 1787-1797; and J. M. Streitz et al., “Endoscopic surveillance of Barrett's esophagus. Does it help?” Journal of Thoracic and Cardiovascular Surgery, 1993, Vol. 105, pp. 383-388. The optimal surveillance and screening strategies for BE are discussed, but many cost-effectiveness analyses focus on frequency and costs of endoscopy as key determinants, as described in J. W. van Sandick et al., “Impact of endoscopic biopsy surveillance of Barrett's oesophagus on pathological stage and clinical outcome of Barrett's carcinoma,” Gut, 1998, Vol. 43(2), pp. 216-22; J. M. Inadomi et al., “Screening and surveillance for Barrett esophagus in high-risk groups: a cost-utility analysis,” Ann Intern Med, 2003, Vol. 138(3), pp. 176-86; D. Provenzale et al., “Barrett's esophagus: a new look at surveillance based on emerging estimates of cancer risk,” Am J Gastroenterol, 1999, Vol. 94(8), pp. 2043-53; and A Sonnenberg et al., “Medical decision analysis of endoscopic surveillance of Barrett's oesophagus to prevent oesophageal adenocarcinoma,” Aliment Pharmacol Ther, 2002, Vol. 16(1), pp. 41-50.

Due to the increasing prevalence of GERD and the medical community's recognition of BE as a risk factor for esophageal cancer, use of endoscopy as a screening and surveillance strategy for BE will increase significantly in the near future. Such increases can incur significant costs to the health care system and to the individual patient. Potential low cost surveillance strategies can include certain endoscopic technologies such as narrow band imaging, chromoendoscopy, or fluorescence endoscopy. Non-endoscopic imaging modalities may also play a role in the management of BE. Methods for directing biopsies to regions of the esophagus containing dysplastic tissue could improve the effectiveness and efficiency of surveillance in patients with BE by increasing surveillance intervals, enabling minimally invasive surgical techniques at an earlier stage of disease progression, or preventing unnecessary interventional procedures.

Provided below is an example of one such optical biopsy technique that can be utilized to obtain information from living human patients.

Optical Coherence Tomography

Optical coherence tomography (OCT) is an optical imaging modality that can use, e.g., a near-infrared light to produce high-resolution (10 μm axial resolution) cross-sectional images of gastrointestinal mucosa. Images may be constructed based on light reflectivity in relation to the properties of the substrate being visualized. OCT techniques can be used to identify structures on a microscopic scale including mucosal layers, “pit and gland” morphology, and glandular structure, as described in S. Brand et al., “Optical coherence tomography in the gastrointestinal tract,” Endoscopy, 2000, Vol. 32(10), pp. 796-803. For example, the OCT techniques can distinguish SIM from squamous fundic, and antral mucosa but can falsely identify gastric cardia as SIM, as discussed in J. M. Poneros et al., “Diagnosis of specialized intestinal metaplasia by optical coherence tomography,” Gastroenterology, 2001, Vol. 120(1), pp. 7-12.

For the OCT techniques to be, e.g., a reliable sensitive and cost effective screening instrument, a characterization of epithelial architecture at the squamocolumnar junction should be accurate enough to distinguish premaligant (SIM) from benign tissue and to identify SIM at the SCJ. Algorithms and methods are required to obtain distinguish SIM from cardia at the SCJ and dysplastic from nonmetaplastic tissue at the gastroesophageal junction.

Described below is an example of the pathologic Haggitt criteria and techniques for diagnosing and grading dysplasia in SIM from H&E stained slides of esophageal biopsies. The Haggitt criteria may be used to aid in rendering a qualitative diagnosis or formulated as a scoring system for semi-quantitative or quantitative diagnosis.

Dysplasia is histologically characterized by, e.g., various degrees and combinations of cytologic atypia and architectural disarray (as described in R. C. Haggit, “Barrett's esophagus, dysplasia, and adenocarcinoma,” Human Pathology, 1994, Vol. 25, pp. 982-93, and E. Montgomery et al. “Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation,” Human Pathology. 2001, Vol. 32, pp. 368-378) for a histologic diagnosis of dysplasia in esophageal SIM. One exemplary set of criteria that can be noted by pathologists and used to render a diagnosis of dysplasia grade is known as the Haggitt criteria. These criteria may be used as part of a scoring system or may be used in a qualitative algorithm for more consistent diagnosis of dysplasia grade. Each of the four Haggitt features are listed below, as described in Montgomery et al. “Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation,” Human Pathology. 2001, Vol. 32, pp. 368-378.

A) Gland Architecture

Glands of dysplastic SIM can be crowded, distorted and irregular in contour with budding, branching, and luminal infoldings. Cribiform glands, cystic dilation, and necrotic debris are more likely to be identified in severe dysplasia.

B) Surface Maturation in Comparison with Underlying Glands

Nondysplastic SIM may have the greatest degree of surface maturation, while HGD can have minimal surface maturation. A high degree of surface maturation implies a low nuclear-to-cytoplasmic ratio at the surface, whereas a low degree of surface maturation indicates a high surface nuclear-to-cytoplasmic ratio.

3) Nuclear Atypia

Cells within dysplastic epithelia generally contain enlarged, hyperchromatic nuclei with irregular nuclear membranes, vesicular (heterogeneous) chromatin, and a loss of nuclear polarity.

4) Inflammation

Inflammation is a confounding factor in the diagnosis of dysplasia since it can independently give rise to distorted glandular architecture and nuclear atypia. Cases where architectural and nuclear atypia may be a result of inflammation are termed indefinite for dysplasia (IND). The interobserver agreement for this diagnosis by histology is low (κ=0.14) (as described in Montgomery et al. “Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation,” Human Pathology. 2001, Vol. 32, pp. 368-378) as it is often reserved for cases where artifacts obscure features required to render a definitive diagnosis or when multiple criteria from different ends of the disease spectrum are simultaneously present.

OBJECTS AND SUMMARY OF THE INVENTION

It is one exemplary object of the present invention is to overcome certain deficiencies and shortcomings of the prior art systems (including those described herein above), and provide an exemplary system, process and software arrangement for interpreting optical biopsy images to provide diagnoses comparable to standard of care histopathology. Another exemplary object of the present invention is to provide exemplary scoring systems and techniques for optical biopsy images. Still another object of the present invention is to provide an exemplary system, process and software arrangement for obtaining histopathologic diagnoses from optical biopsy images.

These and other objects can be achieved by providing an exemplary embodiment of the system, process and software arrangement according to the present invention for obtaining histopathologic diagnoses from optical biopsy images. For example, it is possible to determine a relationship between the properties of the optical biopsy images and images/data utilized in the practice of medicine and pathology to provide a diagnosis. These properties may include, but not limited to a resolution, modes of contrast, spatial and structural features, etc. For example, the resolution of OCT images may be similar to the resolution obtained by a microscope view at low power. Therefore, exemplary architectural features visualized by the OCT systems and methods may be compared to architectural features seen by the conventional histopathology. The contrast in the OCT images may be analogous or similar to conventional histopathologic H&E stains in that high scattering, which may occur at regions of high nuclear content that may be similar or analogous to the basophilic stain of hematoxylin. Furthermore, collagen may have a high scattering observed by the OCT systems and methods, which can be related to linear eosinophilic staining of the lamina propria and submucosa seen in histopathology. These exemplary features of the OCT techniques, systems and method assist with a determination of an analogy between architectural structures seen by this optical biopsy technique and by a low power view of microscopic tissue typically viewed by pathologists.

According to one exemplary embodiment of the present invention, when a relationship is established between the optical biopsy images and histopathology images, criteria and methods used by pathologists to interpret H&E stained slides can be modified on the basis of the comparison between image features and characteristics. Scoring systems and techniques based on these criteria and methods can then be modified and applied to the optical biopsy images themselves. In this manner, a diagnosis can be obtained from the optical biopsy images that are related to conventional histopathologic diagnoses. Advantages of this exemplary embodiment of the system, process, software arrangement and technique can include an ability to utilize prior information obtained from histopathologic correlations between morphology and outcome. Furthermore, since the histopathologic correlations have been formed over decades, the optical biopsy criteria can be similarly reliable for predicting the patient outcome. This can result in a tissue diagnosis from these non- or minimally-invasive methods.

In an exemplary embodiment of the present invention, software systems, arrangements and processes for evaluating an image associated with at least one portion of an anatomical structure are provided. For example, first information associated with the at least one portion of the anatomical structure second information associated with the at least one portion of the anatomical structure can be received. Third information can be generated by determining a relationship between the first information and the second information. Further, the image can be evaluated using a predetermined pathological scoring criteria and the third information.

According to another exemplary embodiment of the present invention, The first information and/or the second information can be associated with a light remitted from the portion of the anatomical structure. The light may be reflected from such portion, and the light can be fluorescence. This portion can be provided in a living subject, and/or may be situated on a microscope slide. The slide can be stained with at least one of Hematoxylin and Eosin, Masson's Trichrome, Papanicolaou's stain, Diff-Qwik, or Periodic Acid Shif.

In still another exemplary embodiment of the present invention, the first information and the second information can be provided for approximately the same location of such portion of the anatomical structure. The third information can be obtained based on physical and chemical structures associated with the first information and the second information. For example, the predetermined pathological scoring criteria can be a Haggitt criteria. The image can be associated with a light remitted from the portion of the anatomical structure. The light may be reflected from such portion, and the light can be fluorescence.

According to a further exemplary embodiment of the present invention, the first information and/or the second information can be obtained by an optical coherence tomography system, a spectrally encoded confocal microscopy system, a confocal microscopy system, a reflectance confocal microscopy system and/or an optical frequency domain imaging system. For example, the anatomical structure may reside below the skin. Further, the slide can be stained with an antibody.

These and other objects, features and advantages of the present invention will become apparent upon reading the following detailed description of embodiments of the invention, when taken in conjunction with the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Further objects, features and advantages of the invention will become apparent from the following detailed description taken in conjunction with the accompanying figures showing illustrative embodiments of the invention, in which:

FIG. 1 is a flow diagram of an exemplary embodiment of a process for generating a scoring system for optical biopsy images by determining relationships between histopathology and optical biopsy data according to the present invention;

FIG. 2 is a flow diagram of another exemplary embodiment of the process for generating the scoring system for the optical biopsy images based on predetermined relationships between the histopathology and the optical biopsy data according to the present invention;

FIG. 3 is a flow diagram of still another exemplary embodiment of the process for generating the scoring system for the optical biopsy images by determining the relationships between the histopathology and the optical biopsy data using a training set of a corresponding optical biopsy and histopathology images;

FIG. 4 is a flow diagram of an exemplary embodiment of a process for generating a tissue diagnosis based on scores from individual criteria, generating a linear combination of said individual scores, and applying a threshold according to the present invention;

FIG. 5A is an OCT image of a non-metaplastic squamous epithelium which shows a horizontally layered architecture of non-metaplasic epithelium;

FIG. 5B is an OCT image of gastric cardia which shows regular, vertical “pit and gland” architecture, a highly scattering epithelial surface, and relatively poor image penetration. Scale bars, 500 μm;

FIG. 6A is an OCT image of specialized intestinal metaplasia (SIM) with a horizontal layered architecture. A. Horizontal layered architecture can be visualized in this OCT image of SIM;

FIG. 6B is an OCT image of specialized intestinal metaplasia (SIM) with a layered architecture, and providing a corresponding histology;

FIG. 7A is an OCT image of SIM without a layered or regular “pit and gland” architecture, low superficial epithelial reflectivity, and relatively good image penetration that are characteristic of SIM at a squamocolumnar junction (SCJ).

FIG. 7B is an OCT image of SIM without the layered architecture, and providing a corresponding histology;

FIG. 8 is a flow diagram of a n exemplary embodiment of the process for differentiating SIM at the SCJ according to the present invention;

FIG. 9A is an OCT image of SIM without dysplasia demonstrates a glandular architecture with a relatively low reflectivity;

FIG. 9B is an OCT image of SIM without dysplasia demonstrates a glandular architecture with a relatively low reflectivity, and that provides a corresponding histology with respect to the image of FIG. 9A with inset demonstrating a low nuclear to cytoplasm ratio in the superficial epithelium;

FIG. 9C is an OCT image of IMC/HGD which enables a visualization of large and irregular dilated glands;

FIG. 9D is an OCT image of irregular, dilated glands that are also shown in the corresponding histology in FIG. 9C;

FIG. 9E is an OCT image of IMC/HGD showing a disorganized architecture and increased surface reflectivity; and

FIG. 9F is an OCT image of SIM, and providing a corresponding histology for the image of FIG. 9E which demonstrates abnormal glandular architecture and an increased superficial nuclear to cytoplasm ratio.

Throughout the figures, the same reference numerals and characters, unless otherwise stated, are used to denote like features, elements, components or portions of the illustrated embodiments. Moreover, while the present invention will now be described in detail with reference to the figures, it is done so in connection with the illustrative embodiments.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENT(S)

FIG. 1 shows an exemplary embodiment of a process for generating an optical biopsy scoring system/procedure for rendering a diagnosis from the optical biopsy images in according to the present invention. This exemplary process shown in FIG. 1 includes taking a set of optical biopsy images in step 100 and obtaining excisional biopsy stained histopathology slides or images thereof in step 110, and determining the features common to both (step 120). The features can be morphologic features determined by correlating the two image sets. The features may comprise individual structures, patterns, intensities that can be identified in both the optical biopsy images and histopathology or alternatively, an interpretation of the optical biopsy image structure based on corresponding histopathologic structures. Examples of the features can include epithelial architecture, epithelial layers, glands, gland shapes, irregular gland features, epithelial maturation, nuclear densities, or the like.

Relationships between the optical biopsy features and the histopathology features are then determined and/or identified in step 130. Once the relationships are determined/identified, then (in step 135) at least one of histopathologic criteria, algorithms, procedures, or scoring systems can be obtained, and in step 140, applied to new optical biopsy images based on predetermined relationships obtained in step 130. In this manner histopathologic scoring systems may be used to render a tissue diagnosis in step 150 based from new optical biopsy images.

FIG. 2 depicts a flow diagram of an exemplary embodiment of a process for determining relationships between optical biopsy images and conventional medical/pathologic diagnoses to obtain an optical biopsy scoring system according to the present invention. In this embodiment, the relationships between histopathology and optical biopsy images can be determined based on a predetermined physical understanding of the contrast methods, resolutions, and/or features that generate the images (step 200). This knowledge may be based on physical principles known in the art or determined by modeling and/or experimentation. For example, it is known that nuclei have a high signal in both OCT and confocal microscopy images. Thus, a predetermined relationship may exist between high OCT and confocal signals and nuclear density. Histopathology images that show a high nuclear to cytoplasmic ratio, indicative of dysplasia for example, should therefore have a high OCT and/or confocal signal intensity. Other relationships known in the art include a high scattering signal from a) collagen, tissue macrophages, b) melanin, c) areas of increased cellular density and a low scattering signal from 1) extracellular matrix, d) cytoplasm, e) the interior of glands, and the like.

With the determination of these predetermined relationships between histopathology and optical biopsy signal content in step 200, relationships between the optical biopsy features and the histopathology features can then be determined in step 210. Once the relationships are determined, then histopathologic criteria, algorithms, and/or scoring systems can be provided in step 220 and applied to new optical biopsy images based on predetermined relationships in step 230 using the relationships obtained in step 220. In this manner histopathologic scoring systems may be used to render a tissue diagnosis based from new optical biopsy images in step 240.

FIG. 3 shows a flow diagram of another exemplary embodiment of a process for generating the scoring system according to the present invention. In this exemplary embodiment, a set of optical biopsy images can be obtained along with corresponding histopathology images from slides that may be obtained at, for example, the same location (step 300). In step 310, the relationships can be identified between optical biopsy images and the histopathology images. The image data sets can be compared and the criteria may be developed in step 320 based on the relationships, including structural, patterns, intensity, between the two data sets. In step 330, histopathologic scoring systems may then be utilized in conjunction with these criteria to develop an optical biopsy scoring system. Alternatively, new optical biopsy scoring system parameters may be generated that are independent of the histopathologic scoring system. The optical biopsy scoring system can then be applied to new optical biopsy images to render a tissue diagnosis in step 350.

In these exemplary embodiments described above, an exemplary generation of the scoring system has been described. FIG. 4 shows a flow diagram of an exemplary embodiment of a process for generating a tissue diagnosis based on the scores according to the present invention. For the scoring systems, a total or final score may be generated in step 420 by adding the individual scores in step 400 for the individual features and/or criteria. The scores may be added linearly and/or may be a linear combination of weighted scores obtained in step 410. A threshold may be placed or applied on the score in step 430 to delineate a certain tissue diagnosis in step 440. Alternatively, in addition to the numeric scores, these exemplary procedures can also be used to generate a flow diagram for a qualitative diagnosis or assist the optical biopsy image reviewer in rendering a qualitative diagnosis.

EXAMPLES

Example 1

Determination of SIM at SCJ from OCT Images

i. Exemplary Design

An exemplary study for the exemplary embodiments of the present invention was a blinded, prospective trial. Its primary objective was to identify OCT image features for differentiating intestinal metaplasia at the SCJ. Patients undergoing a routine outpatient upper endoscopy were requested to participate in the study. OCT images of the SCJ were obtained during the endoscopy procedure. Two pathologists reviewed each biopsy specimen, and noted the presence of the following tissue types: gastric or oxyntic cardia, squamous mucosa, pancreatic metaplasia. The existence of intestinal metaplasia was noted by the presence of goblet cells. Image features of intestinal metaplasia were determined by creating and reviewing an OCT atlas “training set”, containing biopsy-correlated images of known tissue types. These features were then prospectively applied to a “validation set” of unknown tissue types. The sensitivity, specificity, and reproducibility of the image criteria for diagnosis of intestinal metaplasia were determined.

ii. Exemplary OCT System

The exemplary OCT device which can be utilized for the exemplary embodiment of the present invention and used in the study is described in J. M. Poneros et al., “Diagnosis of specialized intestinal metaplasia by optical coherence tomography,” Gastroenterology, 2001, Vol. 120(1), pp. 7-12, and J. M. Poneros et al., “Optical coherence tomography of the biliary tree during ERCP,” Gastrointest Endosc, 2002, Vol. 55(1), pp. 84-8. For example, the light source center wavelength was provided at 1300 nm, and the optical power incident on the tissue was 5.0 mW. The spectral bandwidth of the source was 70 nm, providing an axial resolution of 10 p.m. The catheter diameter was 2.5 mm. Images were acquired in a linear plane longitudinally with dimensions of 5.5 mm (1000 pixels) in length and 2.5 mm (500 pixels) in depth. During image acquisition frames are recorded at a rate of 2 per second and numbered sequentially for reference. A visible aiming laser coincident with the imaging beam allowed the endoscopist to localize the site of mucosa undergoing image acquisition, facilitating biopsy correlation of the imaged site.

Endoscopy and Subject Recruitment

Recruited subjects included patients undergoing routine upper endoscopy and patients with known short (<1 cm) segment intestinal metaplasia at the gastroesophageal junction. A standard gastroscope (Pentax, Model EG 3470K, Tokyo, Japan) with a 3.8 mm instrument channel was utilized.

Exemplary OCT Imaging

Written informed consent was obtained prior to the procedure. After adequate sedation and oropharyngeal anesthesia was achieved, upper endoscopy was performed. The endoscopist identified the SCJ at gastroesophageal junction or Barrett's segment. An OCT catheter probe was introduced through the instrument channel of the endoscope and advanced to the SCJ. Immediately distal to the SCJ, OCT images were acquired and recorded at the mucosal site marked by the visible aiming beam, where one jumbo biopsy was obtained. OCT frames corresponding to the imaged site were documented. Two biopsy-correlated images per patient were obtained.

Histopathology

The biopsy specimens were placed in 10% formalin, embedded in paraffin, processed routinely, and stained with hemotoxylin and eosin.

Description of Pathology Review

Two pathologists reviewed each biopsy specimen and determined the presence of the following epithelial types: gastric cardia, squamous mucosa, serous pancreatic metaplasia, and specialized intestinal metaplasia. For the purposes of this exemplary study, cardia mucosa and oxyntocardia mucosa were grouped together as gastric cardia.

Exemplary OCT Image Analysis

An image atlas “training set” consisting of twenty randomly selected biopsy-correlated images of SIM and twenty randomly selected biopsy correlated images of other tissue types was created. The atlas of images of the training set were reviewed and diagnostic image criteria for SIM were determined. These criteria were then prospectively applied to a “validation set” comprising the remainder of the data set. All OCT images in the validation set were stripped of identifying information and randomly intermixed.

Results

i. Training Set

Squamous epithelium was distinguished by a layered epithelium without glands. FIG. 5A shows an exemplary OCT image of squamous epithelium which demonstrates a horizontally layered architecture. FIG. 5B shows an exemplary OCT image of gastric cardia shows regular, vertical “pit and gland” architecture, a highly scattering epithelial surface, and relatively poor image penetration. Scale bars, 500 μm. Gastric cardia (shown in FIG. 5B) was characterized by the presence of “pit and gland” morphology, regular surface architecture, the presence of a highly reflecting epithelial surface, or poor image penetration.

FIGS. 6A and 6B shows a further exemplary image generated by the exemplary OCT system and process according to the present invention. For example, FIG. 6A shows the OCT image of specialized intestinal metaplasia (SIM) with a horizontal layered architecture. Glands are present in the superficial layer (shown by arrows 600) that differentiate this tissue from squamous epithelium. FIG. 6B shows such exemplary OCT image with a layered architecture and providing a corresponding histology (H&E, 100×). Scale bars, 500 μm.

In particular, SIM was distinguished by the presence of epithelial glands in layered architecture. In cases without layered architecture or “pit and gland” morphology, irregular surface architecture, lack of a highly reflecting epithelial surface, or good light penetration further differentiated SIM from the columnar epithelium of gastric cardia and ectopic pancreas. FIG. 7A depicts an OCT image of SIM without a layered or regular “pit and gland” architecture, low superficial epithelial reflectivity, and relatively good image penetration that are characteristic of SIM at a squamocolumnar junction (SCJ). FIG. 7B shows an OCT image of SIM without the layered architecture, and providing a corresponding histology (H&E, 40×) Scale bars, 500 μm.

An exemplary embodiment of a diagnostic procedure for identifying SIM at the SCJ can be provided using the above-described image criteria is, a flow diagram is shown in FIG. 8. For example, in step 810, it is determined if the layered architecture is provided. If that is the case, then, it is determined (in step 820) whether the glands are epithelium. If so, then the determination is that it is SIM (step 840); otherwise, the determination is squamous (step 830). If, in step 810, it is determined if the layered architecture is not provided, then it is ascertained whether pits and crypts are provided to the surface (step 850). If so, the determination is that it is SIM (step 840). If in step 850 it is determined that pits and crypts are not provided to the surface, then it is ascertained whether broad regular architecture and Dark crisp line are on epithelium (step 860). If so, the determination is that it is SIM (step 840); otherwise, the determination is squamous (step 830). If, in step 850, it is determined that it is possible for pits and crypts to be provided to the surface, then the determination is that it is SIM (step 840); otherwise the determination is squamous (step 830).

When the exemplary embodiment described above with reference to FIG. 8 was retrospectively applied to the training set, it was 85% sensitive (95% CI, 75%-95%) and 95% specific (95% CI, 88%-100%) for differentiating SIM from nonmetaplastic tissue at the SCJ.

Validation Set

Of the 156 biopsy-correlated images comprising the validation set, 36 were removed due to poor image quality, leaving a total of 120 sites for prospective analysis. Table 1 details the histopathology for the validation set. When two blinded OCT readers applied the diagnostic flow chart (FIG. 8) to the validation set, the algorithm was found to be 81% (95% CI 58%-95%) and 86% (95% CI 65%-97%) sensitive and 60% (95% CI 49%-71%) and 58% (95% CI 48%-68%) specific for a diagnosis of SIM at the SCJ. The agreement between the two readers was good (k=0.63). Table 2 demonstrates the variability and performance of the readers' diagnoses following application of the diagnostic algorithm to the validation set.

TABLE 1
Validation Set Histopathology.
Histology Type        # Analyzed in Validation Set
Intestinal Metaplasia 23
Cardia                10
Oxyntocardia          9
Carditis              17
Squamous Mucosa       6
Squamo-Columnar       19
Columnar*             17
Gastric Body          13
Gastritis             5
Pancreatic tissue     1
Total                 120

TABLE 2
Results by Reader.
b Reader 1 Truth Table
	c Pathologic diagnosis	d
a	b	non metaplastic	intestinal metaplasia	total
Diagnosis	non	60	4	64
from OCT	metaplastic
image	intestinal	38	18	56
	metaplasia
	total	98	22	120
c Reader 2 Truth Table
	f Pathologic diagnosis	g
d	e	non metaplastic	intestinal metaplasia	total
Diagnosis	non	57	3	60
from OCT	metaplastic
image	intestinal	41	19	60
	metaplasia
	total	98	22	120

Example 2

Identifying High Grade Dysplasia and Intramucosal Carcinoma in OCT Images of SIM

i Exemplary Study Design

The exemplary study performed was a blinded, prospective trial. Recruited subjects were patients with BE undergoing routine endoscopic surveillance or confirmatory biopsies for IMC or HGD. OCT images of the Barrett's epithelium were obtained during endoscopy. Biopsy correlated OCT images of the esophagus were viewed and scored by a reader blinded to the tissue diagnosis. For each image the score for surface maturation and gland architecture were summed to establish a “dysplasia index”. Each biopsy specimen was independently reviewed, and a consensus diagnosis was rendered.

ii Exemplary OCT System

The exemplary OCT device which can be utilized for the exemplary embodiment of the present invention and used in the study is described in J. M. Poneros et al., “Diagnosis of specialized intestinal metaplasia by optical coherence tomography,” Gastroenterology, 2001, Vol. 120(1), pp. 7-12, and J. M. Poneros et al., “Optical coherence tomography of the biliary tree during ERCP,” Gastrointest Endosc, 2002, Vol. 55(1), pp. 84-8. For example, the light source center wavelength was 1300 nm, and the optical power incident on the tissue was 5.0 mW. The spectral bandwidth of the source was 70 nm, providing an axial resolution of 10 um. The catheter diameter was 2.5 mm. Images were acquired in a linear plane longitudinally with dimensions of 5.5 mm (1000 pixels) in length and 2.5 mm (500 pixels) in depth. During image acquisition frames are recorded at a rate of 4 per second and numbered sequentially for reference. A visible aiming laser coincident with the imaging beam allowed the endoscopist to localize the site of mucosa undergoing image acquisition, facilitating biopsy correlation of the imaged site.

iii Endoscopy and Subject Recruitment

Informed consent was obtained prior to the subject's procedure. Patients with BE undergoing surveillance endoscopy and subjects with known diagnoses of HGD or IMC being evaluated for photodynamic therapy were recruited. Subjects received routine conscious sedation and oropharyngeal anesthesia. A standard endoscope (Pentax, Model EG 3470K, Tokyo, Japan) with a 3.8 mm instrument channel was used.

iv Exemplary OCT Imaging

After adequate sedation and oropharyngeal anesthesia, upper endoscopy was performed. Once the endoscopist identified the gastroesophageal junction and Barrett's segment, an OCT catheter probe was introduced through the instrument channel of the endoscope and advanced to the Barrett's mucosa. OCT images were acquired and recorded at the mucosal site marked by the focusing beam. OCT frames corresponding to the imaged site were documented. One jumbo biopsy was performed at each imaged site.

v Histopathology

The biopsy specimens were placed in 10% formalin, embedded in paraffin, processed routinely, and stained with hemotoxylin and eosin.

Description of Image Scoring System

vi Surface Maturation Definition

OCT measures the intensity of light returning from within a sample. Samples having a higher heterogeneity of optical index of refraction exhibit stronger optical scattering and therefore a stronger OCT signal. Previous research conducted to measure the optical properties of human tissue has shown that the refractive index of chromatin is significantly different than that of the cytoplasm[23]. This data indicates that the OCT signal will increase with increasing nuclear size and density. Histologically, surface maturation is characterized in part by a decrease in the nuclear-to-cytoplasmic ratio of the epithelium at the surface. Incomplete surface maturation, indicative of dysplasia, may therefore be seen as a high surface OCT signal compared to the subsurface signal as shown in FIGS. 9A-F.

For example, FIG. 9A shows an OCT image of SIM without dysplasia demonstrates a glandular architecture with a relatively low reflectivity. FIG. 9B shows an OCT image of SIM without dysplasia demonstrates a glandular architecture with a relatively low reflectivity, and that provides a corresponding histology with respect to the image of FIG. 9A with inset demonstrating a low nuclear to cytoplasm ratio in the superficial epithelium. FIG. 9C shows an OCT image of IMC/HGD which enables a visualization of large and irregular dilated glands 910. FIG. 9D shows an OCT image of irregular, dilated glands 920 that are also shown in the corresponding histology in FIG. 9C. FIG. 9E shows an OCT image of IMC/HGD showing a disorganized architecture and increased surface reflectivity 930. FIG. 9F depicts an OCT image of SIM, and providing a corresponding histology for the image of FIG. 9E which demonstrates abnormal glandular architecture and an increased superficial nuclear to cytoplasm ratio.

vii Gland Architecture Definition

Glands within OCT images are identified as linear structures with alternating low OCT signal (cytoplasm) and high signal (nuclei and lamina propria) as shown in FIGS. 9A-9F. Dilated glands are seen as poorly scattering voids within the mucosa in these figures. Gland irregularity by OCT may be characterized by irregular size, shape, and distribution of these architectural structures as shown therein.

viii Scoring system

For example, the OCT images were stripped of identifying information and randomly intermixed to create a data pool of images. For the purposes of this, images from biopsies consistent with IMC were included as cases of HGD. Without the review of the histopathologic diagnoses, each OCT image was reviewed and scored in the following categories:

• • A) Surface maturation: 0=surface OCT signal weaker than subsurface OCT signal, 1=surface OCT signal equivalent to subsurface OCT signal, 2=surface OCT signal stronger than subsurface OCT signal
  • B) Gland architecture (0=no irregularity, normal appearing glandular architecture; minimal number of smooth dilated glands; 1=mild irregularity, glands were smaller and more densely packed, or large and irregularly shaped; dilated glands were more frequent and closely spaced; 2=moderate/severe irregularity, glands were branching, and budding; dilated glands were highly asymmetric or contained debris within the gland lumen.

For each image the surface maturation and gland architecture scores were summed to establish a dysplasia index.

ix Exemplary Statistical Analyses

A Spearman correlation coefficient (r) was calculated to compare scores of each OCT determined histopathologic feature (surface maturation, gland architecture, and dysplasia index) to the diagnoses of IMC/HGD and to dysplasia (IMC/HGD, LGD, IGD). The sensitivity and specificity of the dysplasia index for the diagnosis of IMC/HGD and dysplasia (IMC/HGD, LGD, IGD) was calculated. The statistics used SAS software (Statistical Analysis System, SAS Institute Inc.) version 8.0. A value of p≦0.05 was considered statistically significant for a two sided test.

Results

The data set was comprised of 242 biopsy-correlated images from 58 patients. Prior to statistical analysis 65 images were removed due to inadequate image quality. Of the 177 remaining images, 49 corresponded to a diagnosis of IMC/HGD, 15 LGD, 8 IGD, 100 SIM, and 5 gastric mucosa. Of the 65 discarded images, 20 corresponded to a diagnosis of IMC/HGD, 13 LGD, 2 IGD, 29 SIM and 1 gastric mucosa. Table 3 summarizes the distribution of histologic diagnoses comprising the data set and displays the average OCT scores of surface maturation, gland architecture, and dysplasia index.

i Distinguishing IMC/HGD from all Others (LGD, IGD, and SIM)

Table 4 shows the Spearman correlation coefficients between each OCT image feature and a diagnosis of IMC/HGD. There was a positive correlation between each of the features and a diagnosis of IMC/HGD: [surface maturation (r=0.49, p<0.0001), gland architecture (r=0.41, p<0.0001), and dysplasia index (r=0.50, p<0.0001)]. Of the three features, the dysplasia index correlated most highly to IMC/HGD.

Table 5 demonstrates a dysplasia index score ≧2 to be 83.3% (95% CI, 70%-93%) sensitive and 75.0% (95% CI, 68%-84%) specific for a diagnosis of IMC/HGD. FIG. 1 (last page) demonstrates examples of IMC/HGD.

ii Distinguishing Dysplasia (IMC/HGD, IGD, LGD) from SIM

Table 4 shows the Spearman correlation coefficients between each of the OCT determined image feature and a diagnosis of dysplasia. There was a positive correlation between each feature and a diagnosis of dysplasia [surface maturation (r=0.47, p<0.0001), gland architecture (r=0.44, p<0.0001), and dysplasia index (r=0.50, p<0.0001)]. The image feature with the highest correlation with dysplasia was the dysplasia index. Table 6 demonstrates a dysplasia index score >2 to be 72.0% (95% CI, 58%-80%) sensitive and 81.0% (95% CI, 72%-88%) specific for a diagnosis of dysplasia.

TABLE 3
Mean OCT scores by histopathology
	Mean scores of image feature as
	determined by OCT
Histopathology		Surface	Gland	Dysplasia
Diagnosis	# in data set	maturation	architecture	index
HGD	49	1.31	1.14	2.45
LGD	15	0.73	0.87	1.60
IGD	8	0.63	0.75	1.38
SIM	100	0.34	0.43	0.77
Gastric	5	0.00	0.20	0.20
Total	177

Of the 49 “HGD” diagnoses, 17 were actually IMC by consensus reads.
14/17 IMC case of had scores >/=2 average (our optimum cutoff) and average dysplasia index score 2.53
26/32 true HGD cases had scores >/=2, and the average dysplasia index score 2.40.

TABLE 4
Correlation between OCT scores and histology
	(Spearman correlation coefficient, r)
	OCT determined histologic feature
	Surface	Gland	Dysplasia
Diagnosis	Maturation	Architecture	Index
HGD	r value	0.48	0.41	0.50
vs all	p value	<0.0001	<0.0001	<0.0001
Dysp	r value	0.47	0.44	0.50
vs SIM	p value	<0.0001	<0.0001	<0.0001

TABLE 5
Truth Table for a diagnosis of HGD using dysplasia index
Dysplasia Index	Sensitivity %	Specificity %
>0	100	0
≧1	88	49
≧2	83	75
≧3	63	87
=4	13	97

TABLE 6
Truth Table diagnosis of dysplasia using dysplasia index
Dysplasia Index	Sensitivity %	Specificity %
≧0	100	0
≧1	82	54
≧2	72	81
≧3	49	88
=4	13	99

The foregoing merely illustrates the principles of the invention. Various modifications and alterations to the described embodiments will be apparent to those skilled in the art in view of the teachings herein. Indeed, the arrangements, systems and methods according to the exemplary embodiments of the present invention can be used with any OCT system, OFDI system, SD-OCT system or other imaging systems, and for example with those described in International Patent Application PCT/US2004/029148, filed Sep. 8, 2004, U.S. patent application Ser. No. 11/266,779, filed Nov. 2, 2005, and U.S. patent application Ser. No. 10/501,276, filed Jul. 9, 2004, the disclosures of which are incorporated by reference herein in their entireties. It will thus be appreciated that those skilled in the art will be able to devise numerous systems, arrangements and methods which, although not explicitly shown or described herein, embody the principles of the invention and are thus within the spirit and scope of the present invention. In addition, to the extent that the prior art knowledge has not been explicitly incorporated by reference herein above, it is explicitly being incorporated herein in its entirety. All publications referenced herein above are incorporated herein by reference in their entireties.

Claims

1-23. (canceled)

24. A process for evaluating at least one image associated with at least one portion of an anatomical structure, comprising:

receiving information associated with the at least one portion of the anatomical structure; and

with a computer processing device, evaluating the at least one image using a predetermined pathological scoring criteria and the information, wherein the predetermined pathological scoring criteria includes at least one of (i) a surface maturation score measured by patterns of reflectance of nuclear density or (ii) a glandular irregularity score.

25. The process according to claim 24, wherein the information is associated with a light remitted from the at least one portion.

26. The process according to claim 25, wherein the light is reflected from the at least one portion.

27. The process according to claim 25, wherein the light is fluorescence.

28. The process according to claim 24, wherein the at least one portion is provided in a living subject.

29. The process according to claim 24, wherein the at least one portion of the anatomical structure is situated on a microscope slide.

30. The process according to claim 29, wherein the slide is stained with at least one of Hematoxylin and Eosin, Masson's Trichrome, Papanicolaou's stain, Diff-Qwik, or Periodic Acid Shiff.

31. The process according to claim 24, wherein the information is provided for approximately the same location of the at least one portion of the anatomical structure.

32. The process according to claim 24, wherein the information is obtained based on physical and chemical structures associated with the information.

33. The process according to claim 24, wherein the image is associated with a light remitted from the at least one portion.

34. The process according to claim 33, wherein the light is reflected from the at least one portion.

35. The process according to claim 33, wherein the light is fluorescence.

36. The process according to claim 33, wherein the at least one portion is provided in a living subject.

37. The process according to claim 24, wherein the first information is obtained by an optical coherence tomography system.

38. The process according to claim 24, wherein the information is obtained by a spectrally encoded confocal microscopy system.

39. The process according to claim 24, wherein the anatomical structure resides below the skin.

40. The process according to claim 39, wherein the information is obtained by a confocal microscopy system.

41. The process according to claim 40, wherein the information is obtained by a reflectance confocal microscopy system.

42. The process according to claim 24, wherein the information is obtained by an optical frequency domain imaging system.

43. The process according to claim 29, wherein the slide is stained with an antibody.

44. An apparatus for evaluating at least one image associated with at least one portion of an anatomical structure, comprising:

a computer processing device which is configured to:

a) receiving information associated with the at least one portion of the anatomical structure; and

b) evaluating the at least one image using a predetermined pathological scoring criteria and the information, wherein the predetermined pathological scoring criteria includes at least one of (i) a surface maturation score measured by patterns of reflectance of nuclear density or (ii) a glandular irregularity score.

45. A software system for evaluating at least one image associated with at least one portion of an anatomical structure, the system being provided on a hardware storage arrangement and executed by a computer processing device, the software system comprising:

a first set of instructions being stored on a non-transitory computer-readable medium and which, when execute by the computer processing device, configures the computer processing device to information associated with the at least one portion of the anatomical structure; and

a second set of instructions being stored on the non-transitory computer-readable medium and which, when execute by the computer processing device, configures the computer processing device to evaluating the at least one image using a predetermined pathological scoring criteria and the information, wherein the predetermined pathological scoring criteria includes at least one of (i) a surface maturation score measured by patterns of reflectance of nuclear density or (ii) a glandular irregularity score.