Abstract: A dynamic image processing apparatus includes: a hardware processor that: extracts a lung-field region from at least one of a plurality of frame images of a chest dynamic image obtained by radiographing a dynamic state of a chest of an examinee; sets a feature point in a position that moves according to a movement of a lung field due to respiration in the lung-field region extracted by the hardware processor; searches a frame image other than a frame image in which the feature point has been set for a corresponding point that corresponds to the feature point set by the hardware processor, and estimates a correspondence relationship of each pixel in the lung-field region among the plurality of frame images in accordance with a positional relationship between the feature point set by the hardware processor and the corresponding point searched for by the hardware processor.

Abstract: The embodiments of the present invention provide training and construction methods and apparatus of a neural network for object detection, an object detection method and apparatus based on a neural network and a neural network. The training method of the neural network for object detection, comprises: inputting a training image including a training object to the neural network to obtain a predicted bounding box of the training object; acquiring a first loss function according to a ratio of the intersection area to the union area of the predicted bounding box and a true bounding box, the true bounding box being a bounding box of the training object marked in advance in the training image; and adjusting parameters of the neural network by utilizing at least the first loss function to train the neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a whitened neural network layer. One of the methods includes receiving an input activation generated by a layer before the whitened neural network layer in the sequence; processing the received activation in accordance with a set of whitening parameters to generate a whitened activation; processing the whitened activation in accordance with a set of layer parameters to generate an output activation; and providing the output activation as input to a neural network layer after the whitened neural network layer in the sequence.

Abstract: Embodiments of the present invention provide a method and a device for three-dimensional feature-embedded image object component-level semantic segmentation, the method includes: acquiring three-dimensional feature information of a target two-dimensional image; performing a component-level semantic segmentation on the target two-dimensional image according to the three-dimensional feature information of the target two-dimensional image and two-dimensional feature information of the target two-dimensional image. In the technical solution of the present application, not only the two-dimensional feature information of the image but also the three-dimensional feature information of the image are taken into consideration when performing the component-level semantic segmentation on the image, thereby improving the accuracy of the image component-level semantic segmentation.

Abstract: Systems and methods are described to generate, using a first image generation technique, a first image based on the first image data, display the first image to an operator, receive, from the operator, one or more indications of features in the first image of the patient volume, generate, using a second image generation technique, a second image based on the first image data, perform automated feature extraction on the second image to automatically extract information associated with features of the patient volume, and output a feature report of the patient volume based on the one or more indications of features and the information associated with features.

Abstract: Techniques related to implementing convolutional neural networks for object recognition are discussed. Such techniques may include generating a set of binary neural features via convolutional neural network layers based on input image data and applying a strong classifier to the set of binary neural features to generate an object label for the input image data.

Abstract: A method of designing or selecting an implantable medical device comprises the steps of: i) obtaining a plurality of measured data points of a characteristic of an anatomical feature of an individual; ii) using said data points to construct a surrogate model of said characteristic, the surrogate model being constructed by interpolating or regressing measured data points of the characteristic, and using said surrogate model to obtain predicted values of said characteristic at a plurality of locations; iii) using said predicted values to determine or select at least one value of a design parameter of the implantable medical device. There is further disclosed a method of monitoring or diagnosing a disease or disorder.

Abstract: A living body state estimation apparatus acquires information indicating a state of a living body. The living body state estimation apparatus is configured to include an electrocardiogram signal acquisition unit which acquires an electrocardiogram signal of the living body and an information acquisition unit which acquires a parameter as the information, the parameter specifying a predetermined function indicating a probability distribution for a reference wave interval which is a time interval between peaks of consecutive predetermined reference waves in the acquired electrocardiogram signal.

Abstract: A computer-readable storage medium may be configured to store a program comprising instructions configured to, when executed by a computing device, cause the computing device to detect a selection of a partial area of the image, transform the image into a transformed image in which the selected partial area is positioned in a center of the transformed image, extract at least one feature from the transformed image, using a deep learning technique, enhance at least one feature of the at least one extracted feature, restore, as a restored image, at least one feature of the at least one enhanced feature, and inversely transform the restored image to provide segmented images.

Abstract: A wind park comprising at least two wind turbines that produce electrical power by means of a wind rotor and a generator and delivers this to an accumulating network, and comprising a park master that is configured to control said wind turbines and has a power regulator whose input is supplied with a target power signal and, at whose output, power control signals are emitted for the wind turbines, said power regulator comprising a feed-forward control module that imposes a value for the target power onto the output of said power regulator by means of a multiplication element.

Abstract: A three dimensional bounding box is determined from a two dimensional image. A two dimensional bounding box is calculated based on a detected object within the image. A three dimensional bounding box is parameterized as having a yaw angle, dimensions, and a position. The yaw angle is defined as the angle between a ray passing through a center of the two dimensional bounding box and an orientation of the three dimensional bounding box. The yaw angle and dimensions are determined by passing the portion of the image within the two dimensional bounding box through a trained convolutional neural network. The three dimensional bounding box is then positioned such that the projection of the three dimensional bounding box into the image aligns with the two dimensional bounding box previously detected. Characteristics of the three dimensional bounding box are then communicated to an autonomous system for collision and obstacle avoidance.

Abstract: An image reporting method is provided. The image reporting method comprises the steps of retrieving an image representation of a sample structure from an image source; mapping a generic structure to the sample structure, the generic structure being related to the sample structure; determining a region of interest within the sample structure based on content of the image representation of the sample structure; providing a focused set of representations of diagnostic knowledge which is contextually appropriate to the region of interest and prompting the user to select at least one diagnostic finding from the focused set of knowledge representation or by entering free-form text; and generating a diagnostic report based on the selections and free-form text entries.

Abstract: According to an aspect of an exemplary embodiment, an apparatus for avoiding region of interest (ROI) re-detection includes a detector configured to detect an ROI from an input medical image; a re-detection determiner configured to determine whether the detected ROI corresponds to a previously-detected ROI using pre-stored user determination information; and an ROI processor configured to perform a process for the detected ROI based on the determination.

Abstract: The present invention provides an energy based dissection device that automatically provides a nerve protection function. Specifically, the present invention operatively connects nerve monitoring technology and energy based dissection technology to provide a device that provides energy based dissection functionality that cannot operate, or operates differently, upon receipt of real time information from the nerve monitoring functionality that nerve damage may be imminent in the absence of such safety shutdown. The present invention also creates a real-time graphical display of the nerve, including size and location relative to the energy based dissection device to enable the operator to safely and accurately avoid damaging the nerve. Accordingly, the present invention provides a surgical device that removes human error and reaction time issues which prevent unintended dissection and concomitant nerve damage.

Abstract: Techniques for increasing robustness of a convolutional neural network based on training that uses multiple datasets and multiple tasks are described. For example, a computer system trains the convolutional neural network across multiple datasets and multiple tasks. The convolutional neural network is configured for learning features from images and accordingly generating feature vectors. By using multiple datasets and multiple tasks, the robustness of the convolutional neural network is increased. A feature vector of an image is used to apply an image-related operation to the image. For example, the image is classified, indexed, or objects in the image are tagged based on the feature vector. Because the robustness is increased, the accuracy of the generating feature vectors is also increased. Hence, the overall quality of an image service is enhanced, where the image service relies on the image-related operation.

Abstract: A method for tracking three-dimensional movement of an object and focusing thereon is provided. Software repeatedly detects three-dimensional locations of one or more objects and image quality by sampling images at many parameter settings as the objects move in three-dimensional environment. The images are ranked occurring to a fitness score and parameter settings for subsequent images to established using a biologically-inspired algorithm.

Abstract: A magnetic resonance imaging apparatus includes a trigger generating unit, a blood flow image generating unit and a control unit. The trigger generating unit acquires blood flow information of an object and generates a trigger based on the blood flow information. The blood flow image generating unit acquires imaging data with using the trigger and generates blood flow image data. The control unit controls so as to repeatedly perform a probe sequence for acquiring the blood flow information and an imaging sequence for acquiring the imaging data alternately.

Abstract: Systems and methods are disclosed for providing a cardiovascular score for a patient. A method includes receiving, using at least one computer system, patient-specific data regarding a geometry of multiple coronary arteries of the patient; and creating, using at least one computer system, a three-dimensional model representing at least portions of the multiple coronary arteries based on the patient-specific data. The method also includes evaluating, using at least one computer system, multiple characteristics of at least some of the coronary arteries represented by the model; and generating, using at least one computer system, the cardiovascular score based on the evaluation of the multiple characteristics. Another method includes generating the cardiovascular score based on evaluated multiple characteristics for portions of the coronary arteries having fractional flow reserve values of at least a predetermined threshold value.

Abstract: Described are systems, media, and methods for applying deep convolutional neural networks to medical images to generate a real-time or near real-time diagnosis.

Abstract: A method and system acquiring, processing and displaying breast ultrasound images in a way that makes breast ultrasound screening more practical and thus more widely used, and reduces missing cancers in screening and diagnosis, using automated scanning of chestwardly compressed breasts with ultrasound. Enhanced, whole-breast navigator overview images are generated from scanning breasts with ultrasound that emphasize abnormalities in the breast while excluding obscuring influences of non-breast structures, particularly those external to the breast such as ribs and chest wall, and differentiating between likely malignant and likely benign abnormalities and otherwise enhancing the navigator overview image and other images, to thereby reduce the time to read, screen, or diagnose to practical time limits and also reduce screening or diagnostic errors.

Abstract: A computerized method of adapting a presentation of ultrasonographic images during an ultrasonographic fetal evaluation. The method comprises performing an analysis of a plurality of ultrasonographic images captured by an ultrasonographic probe during an evaluation of a fetus, automatically identifying at least one location of at least one anatomical landmark of at least one reference organ or tissue or body fluid of the fetus in the plurality of ultrasonographic images based on an outcome of the analysis, automatically localizing a region of interest (ROI) in at least some of the plurality of ultrasonographic images by using at least one predefined locational anatomical property of the at least one anatomical landmark, and concealing the ROI in a presentation of the at least some ultrasonographic images during the evaluation. At least one anatomical landmark is imaged in the presentation and not concealed by the ROI.

Abstract: In a combination invasive and non-invasive bioparameter monitoring device an invasive component measures the bioparameter and transmits the reading to the non-invasive component. The non-invasive component generates a bioparametric reading upon insertion by the patient of a body part. A digital processor processes a series over time of digital color images of the body part and represents the digital images as a signal over time that is converted to a learning vector using mathematical functions. A learning matrix is created. A coefficient of learning vector is deduced. From a new vector from non-invasive measurements, a new matrix of same size and structure is created. Using the coefficient of learning vector, a recognition matrix may be tested to measure the bioparameter non-invasively. The learning matrix may be expanded and kept regular. After a device is calibrated to the individual patient, universal calibration can be generated from sending data over the Internet.

Abstract: A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches.

Abstract: For therapy response assessment, texture features are input for machine learning a classifier and for using a machine learnt classifier. Rather than or in addition to using formula-based texture features, data driven texture features are derived from training images. Such data driven texture features are independent analysis features, such as features from independent subspace analysis. The texture features may be used to predict the outcome of therapy based on a few number of or even one scan of the patient.

Abstract: Computer-implemented methods of improved quantitative interpretation of cell index profiles are provided.

Abstract: The present invention relates to an apparatus and a method for void size determination of voids within an object into which an aerosol containing magnetic particles has been introduced, in particular for determining the size of a patient's pulmonary alveoli, said patient having inhaled an aerosol containing magnetic particles To review information concerning the lung structure, it is proposed to use magnetic particle imaging. First and second detection signals are acquired subsequently at different moments in time after introduction of the aerosol containing the magnetic particles into the object, in particular after inhalation of the aerosol by the patient. These detection signals are exploited, in particular the drop in intensity and/or the signal decay time, to get information about the diffusion of the magnetic particles within the voids, in particular alveoli, and to retrieve information therefrom about the size of the voids, in particular alveoli.

Abstract: A method of assessing stenosis severity for a patient includes obtaining patient information relevant to assessing severity of a stenosis, including anatomical imaging data of the patient. Based on the anatomical imaging data, the existence of any lesions of concerns may be identified. A three dimensional image can be generated of any irregular shaped lesion of concern and a surrounding area from the patient anatomical imaging data. A plurality of comparative two dimensional lesion specific models may be created that have conditions that correspond to the three dimensional model. The comparative two dimensional models may represent vessels having regular shaped lesions with each of the comparative two dimensional models represents a different stenosis severity. The three dimensional model can then be mapped to one of the plurality of comparative two dimensional models. After this mapping, a diagnosis of whether the patient has coronary artery disease may be made.

Abstract: A method for detecting polyps in endoscopy images includes pruning a plurality of two dimensional digitized images received from an endoscopy apparatus to remove images that are unlikely to depict a polyp, where a plurality of candidate images remains that are likely to depict a polyp, pruning non-polyp pixels that are unlikely to be part of a polyp depiction from the candidate images, detecting polyp candidates in the pruned candidate images, extracting features from the polyp candidates, and performing a regression on the extracted features to determine whether the polyp candidate is likely to be an actual polyp.

Abstract: Systems and techniques for delivery and medical support are disclosed herein. In some embodiments, a drone delivery system may include receiving logic and communication logic. The receiving logic may be configured to receive a request signal indicative of a package request event proximate to a target device, wherein the request signal comprises sensor data indicative of conditions proximate to the target device or a request signal transmitted to the receiving device from the target device. The communication logic may be configured to instruct a drone to carry a package to the target device in response to the request signal. The drone may be configured to perform an environmental scan during transit to adjust a route to the target device. Other embodiments may be disclosed and/or claimed.

Abstract: In alternative embodiments, the invention provides compositions, products of manufacture and medical devices, and methods, using optical coherence tomography to monitor for a physiological event and/or a state prior to the physiological event. The invention also provides computer program products and computer implemented methods to capture, analyze, and display optical coherence tomography images of neural tissue to measure, detect and/or monitor for a physiological event and/or a state prior to the physiological event.

Abstract: A method for determining functional severity of a stenosis includes: (a) generating a simulated perfusion map from a calculated blood flow; (b) comparing the simulated perfusion map to a measured perfusion map to identify a degree of mismatch therebetween, the measured perfusion map representing perfusion in a patient; (c) modifying a parameter in a model used in calculating the blood flow when the degree of mismatch meets or exceeds a predefined threshold; (d) computing a hemodynamic quantity from the simulated perfusion map when the degree of mismatch is less than the predefined threshold, the hemodynamic quantity being indicative of the functional severity of the stenosis; and (e) displaying the hemodynamic quantity. Systems for determining functional severity of a stenosis are described.

Abstract: A monitoring system includes a Doppler radar formed with an IEEE 802 protocol transmitter and an IEEE 802 protocol receiver to detect motion, wherein frequency waves are transmitted by the IEEE 802 protocol transmitter and reflected waves are received by the IEEE protocol 802 receiver; and an analyzer to perform Doppler operations using the IEEE 802 protocol transmitter and receiver, wherein the analyzer calibrates a training Doppler radar signal during a training phase to develop a model and wherein the analyzer applies the model with the Doppler radar signal during an operational phase to determine position.

Abstract: A method automatically scores calcium in the aorta and other arteries of the body using calcium plaque definitions that include subject specific in vivo blood/muscle density measurements, subject specific voxel statistical parameters and 2D and 3D voxel connectivity criteria to automatically identify the plaques. The images are optionally calibrated with external phantoms or internal reference tissue. Aortic calcium is identified automatically without manual marking. Potential false plaques from bone are automatically excluded. A 3D coordinate system provides the specific coordinates of the detected plaques, which are displayed in a plaque map for follow-up exams or ease in plaque review.

Abstract: Methods and systems for in vivo classification of tissue are disclosed. The tissue is irradiated with light from multiple light sources and light scattered and fluoresced from the tissue is received. Distinct emissions of the sample are identified from the received light. An excitation-emission matrix is generated (1002). On-diagonal and off-diagonal components of the excitation-emission matrix are identified (1004, 1006, 1008). Spectroscopic measures are derived from the excitation-emission matrix (1014), and are compared to a database of known spectra (1016) permitting the tissue to be classified as benign or malignant (1018). An optical biopsy needle or an optical probe may be used to contemporaneously classify and sample tissue for pathological confirmation of diagnosis.

Abstract: A method and apparatus is described for optically scanning a field of view, the field of view including at least part of an organ as exposed during surgery, and for identifying and classifying areas of tumor within the field of view. The apparatus obtains a spectrum at each pixel of the field of view, and classifies pixels with a kNN-type or neural network classifier previously trained on samples of tumor and organ classified by a pathologist. Embodiments use statistical parameters extracted from each pixel and neighboring pixels. Results are displayed as a color-encoded map of tissue types to the surgeon. In variations, the apparatus provides light at one or more fluorescence stimulus wavelengths and measures the fluorescence light spectrum emitted from tissue corresponding to each stimulus wavelength. The measured emitted fluorescence light spectra are further used by the classifier to identify tissue types in the field of view.

Abstract: Methods and apparatus for classifying tissue use features of Raman spectra and background fluorescent spectra. The spectra may be acquired in the near-infrared wavelengths. Principal component analysis and linear discriminant analysis of reference spectra may be used to obtain a classification function that accepts features of the Raman and background fluorescence spectra for test tissue and yields an indication as to the likelihood that the test tissue is abnormal. The methods and apparatus may be applied to screening for skin cancers or other diseases.

Abstract: When detecting and classifying hypo-metabolic regions in the brain to facilitate dementia diagnosis, a patient's brain scan image, generated using an FDG-PET scan, is compared to a plurality of hypo-metabolic region patterns in brain scan images associated with a plurality of types of dementia. In a fully automated mode, the patient's scan is compared to all scans stored in a knowledge base, and a type of dementia associated with a most likely match is output to a user along with a highlighted image of the patient' s brain. In a semi-automated mode, a user specifies two or more types of dementia, and the patient's scan is compared to scans typical of the specified types. Diagnosis information including respective likelihoods for each type is then output to the user. Additionally, the user can adjust a threshold significance level to increase or decrease a number of voxels that are included in hypo-metabolic regions highlighted in the patient' brain scan image.

Abstract: A method of contoured-anatomy dose repositioning (CADR) as a means to automatically reposition a patient to better recover the planned dose distribution without reoptimizinq the treatment plan. CADR utilizes planning CT images, the planned dose distribution, and on-line images for repositioning dose distribution on a given day. Contours are also placed upon the images using manual, automatic, template-based, or other techniques. CADR then optimizes the rigid-body repositioning of the patient so that the daily dose distribution closely matches the planned dose distribution.

Abstract: A method for measuring ventricular dimensions from M-mode echocardiograms, includes providing a digitized M-mode echocardiogram image, running a plurality of local classifiers, where each local classifier trained to detect a landmark on either an end-diastole (ED) line or an end-systole (ES) line in the image, recording all possible landmarks detected by the classifiers, where a search range in an N-dimensional parameter space defined by the landmarks for each dimension is reduced to a union of subsets, where each dimension of the parameter space corresponds a landmark, for each combination of possible landmarks, checking if an order of the landmarks is consistent with a known ordering of the landmarks, and if the order is consistent, running a global detector on each consistent combination of landmarks to find a landmark combination with a highest detection probability as a confirmed landmark detection, where the landmarks are used for measuring ventricular dimensions.

Abstract: A combination of a haptic device and a computer-assisted medical system is used for interactive haptic positioning of a medical device coupled to the haptic device. A reconfigurable haptic object facilitates the positioning of the medical device and/or the haptic device. The haptic object may be modified in response to application of a force against the haptic object by a user of the haptic device pushing the haptic device against the haptic object. Preferably, the haptic object moves in the direction of the force applied by the haptic device. The medical device may be guided to a desired pose relative to a target area from its current position. The user may approach the target area from its current position and still be provided with haptic cues to enable the user to guide the medical device to the target area.

Abstract: A method for creating a model of a part of the anatomy from the scan data of several subjects is described. The method comprises the steps of collecting scan data; applying a feature detector to the scan data; converted the output of the feature detector into a common reference system; and accumulating the converted data to generate the model. It is therefore possible for the method to generate a model from the scan data of several subjects automatically. The method may also include an optional step of receiving user input to select which of the accumulated data should be included in the final model. This user input requires much less effort than manual contouring and is substantially independent of the number of subjects used to create the model.

Abstract: The present invention describes a method and system for intelligent diagnostic relevant information processing and analysis. Information associated with a patient is processed via an image reading platform. Based on such processed information, a matrix of diagnosis decisions containing diagnostic related information is generated via a matrix of diagnosis decision platform. A diagnostic decision is made based on the diagnostic relevant information. The image reading platform and/or the matrix of diagnosis decision platform encapsulate information and toolkits to be used to manipulate the information.

Abstract: The invention provides an emotional state determination method capable of quantitatively and accurately measuring an emotional state of a human being without requiring a special measuring environment. At least one of oxy-hemoglobin densities and deoxy-hemoglobin densities in blood of a plurality of measurement regions of a human brain cortex of a subject 1 are measured in time series, respectively, using a near-infrared spectroscopy. Then, cross-correlation coefficients of plural sets of time-variable change data are computed for each time (for each sampling period). Each of the plural sets of the time-variable change data comprises two of the time-variable change data which are selected by permutations and combinations from among at least one of the time-variable change data on the measured oxy-hemoglobin densities and the time-variable change data on the measured deoxy-hemoglobin densities.

Abstract: A decision-support system and computer implemented method automatically measures tee midline shift in a patient's brain using Computed Tomography (CT) images. The decision-support system and computer implemented method applies machine learning methods to features extracted from multiple sources, including midline shift, blood amount, texture pattern and other injury data, to provide a physician an estimate of intracranial pressure (ICP) levels. A hierarchical segmentation method, based on Gaussian Mixture Mode! (GMM), is used. In this approach, first an Magnetic Resonance Image (MRI) ventricle template, as prior knowledge, is used to estimate the region for each ventricle. Then, by matching the ventricle shape it) CT images to fee MRI ventricle template set, the corresponding MRI slice is selected. From the shape matching result, the feature points for midline estimation in CT slices, such as the center edge points of the lateral ventricles, are detected.

Abstract: A system for determining a location of an electrode of a medical device (e.g., a catheter) in a body of a patient includes a localization block for producing an uncompensated electrode location, a motion compensation block for producing a compensation signal (i.e., for respiration, cardiac, etc.), and a mechanism for subtracting the compensation signal from the uncompensated electrode location. The result is a corrected electrode location substantially free of respiration and cardiac artifacts. The motion compensation block includes a dynamic adaptation feature which accounts for changes in a patient's respiration patterns as well as intentional movements of the medical device to different locations within the patient's body. The system further includes an automatic compensation gain control which suppresses compensation when certain conditions, such as noise or sudden patch impedance changes, are detected.

Abstract: A method and apparatus are disclosed that reduces variation in radiation therapy treatment planning among plurality of users within the same or different geographic locations. The system includes a method and an apparatus that provide users with the knowledge information and utilizing the knowledge information to contour target volumes for radiation treatment planning. The mode of operation includes utilizing a stand-alone workstation or a server computer connected to the plurality of thin client workstations.

Abstract: The systems and methods described herein provide for fast and accurate image segmentation through the application of a multi-stage classifier to an image data set. An image processing system is provided having a processor configured to apply a multi-stage classifier to the image data set to identify a distinctive region. The multi-stage classifier can include two or more component classifiers. The first component classifier can have a sensitivity level configured to identify one or more target regions in the image data set and the second component classifier can have a specificity level configured to confirm the presence of the distinctive region in any identified target regions. Also provided is a classification array having multiple multi-stage classifiers for identification and confirmation of more than one distinctive region or for the application of different classification configurations to the image data set to identify a specific distinctive region.

Abstract: According to one embodiment there is provided a method of selecting a plurality of M atlases from among a larger group of N candidate atlases to form a multi-atlas data set to be used for computer automated segmentation of novel image data sets to mark objects of interest therein. A set of candidate atlases is used containing a reference image data set and segmentation data. Each of the candidate atlases is segmented against the others in a leave-one-out strategy, in which the candidate atlases are used as training data for each other. For each candidate atlas in turn, the following is carried out: registering; segmenting; computing an overlap; computing a value of the similarity measure for each of the registrations; and obtaining a set of regression parameters by performing a regression with the similarity measure being the independent variable and the overlap being the dependent variable.

Abstract: Disclosed is a medical outcome prediction tool that predicts an individual patient's medical outcomes by identifying patients having a same disease; selecting a set of characteristics unique to an individual; determining the similarities between the individual and other cases; and calculating the expected outcome for the individual that is proportional to a weighted sum of outcomes of similar cases. The similarities can be determined by calculating the number of matches between the individual and cases over the set of characteristics, and using that result to determine a similarity score.

Abstract: A method of finding the location of an occluded portion of a blood vessel relative to a three-dimensional angiographic image of a subject's vasculature includes identifying the location of the occluded portion of the blood vessel on each of a series of displayed two dimensional images derived from the three dimensional image data in planes substantially transverse to direction of the occluded portion of the vessel. The identified locations in the occluded portion of the vessel can then be used to determine the path of the occluded portion of the vessel.