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: A method and apparatus for automatically performing medical image analysis tasks using deep image-to-image network (DI2IN) learning. An input medical image of a patient is received. An output image that provides a result of a target medical image analysis task on the input medical image is automatically generated using a trained deep image-to-image network (DI2IN). The trained DI2IN uses a conditional random field (CRF) energy function to estimate the output image based on the input medical image and uses a trained deep learning network to model unary and pairwise terms of the CRF energy function. The DI2IN may be trained on an image with multiple resolutions. The input image may be split into multiple parts and a separate DI2IN may be trained for each part. Furthermore, the multi-scale and multi-part schemes can be combined to train a multi-scale multi-part DI2IN.

Abstract: Systems and methods are provided for generating a synthesized medical image patch of a nodule. An initial medical image patch and a class label associated with a nodule to be synthesized are received. The initial medical image patch has a masked portion and an unmasked portion. A synthesized medical image patch is generated using a trained generative adversarial network. The synthesized medical image patch includes the unmasked portion of the initial medical image patch and a synthesized nodule replacing the masked portion of the initial medical image patch. The synthesized nodule is synthesized according to the class label. The synthesized medical image patch is output.

Abstract: Patient, user, and/or AI information are used in a multi-objective optimization to select one of a plurality of available AIs for a task. On a patient or user-specific basis, an optimal AI is selected and applied for medical imaging or other healthcare actions. The selection may be before application, avoiding costs of applying multiple AIs to obtain the best results. The optimization may be based on statistical feedback from the user for various of the available AIs, providing information not otherwise available. The optimization may be based on AI performance, AI inclusion and/or exclusion criteria, and/or pricing information. By using optimization based on various information related to the patient, user, and/or available AI, the application of AI for a given user and/or patient by the computer may be improved. The computer operates better to provide more focused information through AI application.

Abstract: System and methods are provided for localizing a target object in a medical image. The medical image is discretized into a plurality of images having different resolutions. For each respective image of the plurality of images, starting from a first image and progressing to a last image with the progression increasing in resolution, a sequence of actions is performed for modifying parameters of a target object in the respective image. The parameters of the target object comprise nonlinear parameters of the target object. The sequence of actions is determined by an artificial intelligence agent trained for a resolution of the respective image to optimize a reward function. The target object is localized in the medical image based on the modified parameters of the target object in the last image.

Abstract: Machine learning is used to assess data for a patient in an emergency, providing rapid diagnosis based on a large amount of information. Assistance in triage may be provided. Given the large variety of patients and conditions that may occur, the machine learning may rely on synthetically generated images for more accurate prediction. The machine learning may accurately predict even with missing information and may be used to determine what missing information for a given patient is more or less important to obtain.

Abstract: A computer-implemented method for deriving biopsy results in a non-invasive manner includes acquiring a plurality of training data items. Each training data item comprises non-invasive patient data and one or more biopsy derived scores associated with an individual. The method further includes extracting a plurality of features from the non-invasive patient data based on the one or more biopsy derived scores and training a predictive model to generate a predicted biopsy score based on the plurality of features and the one or more biopsy derived scores.

Abstract: Medical images may be classified by receiving a first medical image. The medical image may be applied to a machine learned classifier. The machine learned classifier may be trained on second medical images. A label of the medical image and a measure of uncertainty may be generated. The measure of uncertainty may be compared to a threshold. The first medical image and the label may be output when the measure of uncertainty is within the threshold.

Abstract: In personalized object creation, for implants, medical imaging is used to derive a model personalized to a patient. The model may be of a dynamic structure, such as part of the cardiovascular system, and is used to print the implant itself. The model may be used to print a mold to create the implant, a scaffold on which to grow tissue, and/or tissue itself. In another or additional approach, the medical imaging information is used to determine tissue properties. Differences in a material property of the anatomy is mapped to different materials used by a multi-material 3D printer, resulting in a printed object reflecting the size, shape, and/or other material property of the anatomy of the patient.

Abstract: A method and system for interactive patient-specific simulation of liver tumor ablation is disclosed. A patient-specific anatomical model of the liver and circulatory system of the liver is estimated from 3D medical image data of a patient. A computational domain is generated from the patient-specific anatomical model of the liver. Blood flow in the liver and the circulatory system of the liver is simulated based on the patient-specific anatomical model. Heat diffusion due to ablation is simulated based on a virtual ablation probe position and the simulated blood flow in the liver and the circulatory system of the liver by solving a bio-heat equation for each node on the level-set representation using a Lattice-Boltzmann method (LBM) implementation. Cellular necrosis in the liver is computed based on the simulated heat diffusion. Visualizations of a computed necrosis region and temperature maps of the liver are generated.

Abstract: Methods and systems for image registration using an intelligent artificial agent are disclosed. In an intelligent artificial agent based registration method, a current state observation of an artificial agent is determined based on the medical images to be registered and current transformation parameters. Action-values are calculated for a plurality of actions available to the artificial agent based on the current state observation using a machine learning based model, such as a trained deep neural network (DNN). The actions correspond to predetermined adjustments of the transformation parameters. An action having a highest action-value is selected from the plurality of actions and the transformation parameters are adjusted by the predetermined adjustment corresponding to the selected action. The determining, calculating, and selecting steps are repeated for a plurality of iterations, and the medical images are registered using final transformation parameters resulting from the plurality of iterations.

Abstract: Systems and methods are described for automatically identifying an anatomical landmark in a medical image according to local preferences associated with a particular clinical site. A medical image for performing a medical procedure is received. An anatomical landmark is identified in the medical image using a pre-trained machine learning algorithm. Feedback relating to the identified anatomical landmark is received from a user associated with a particular clinical site. The feedback is received during a normal workflow for performing the medical procedure. The pre-trained machine learning algorithm is retrained based on the received feedback such that the retrained machine learning algorithm is trained according to local preferences associated with the particular clinical site.

Abstract: A method and system for estimating physiological heart measurements from medical images and clinical data disclosed. A patient-specific anatomical model of the heart is generated from medical image data of the patient. A patient-specific multi-physics computational heart model is generated based on the patient-specific anatomical model by personalizing parameters of a cardiac electrophysiology model, a cardiac biomechanics model, and a cardiac hemodynamics model based on medical image data and clinical measurements of the patient. Cardiac function of the patient is simulated using the patient-specific multi-physics computational heart model. The parameters can be personalized by inverse problem algorithms based on forward model simulations or the parameters can be personalized using a machine-learning based statistical model.

Abstract: Intelligent multi-scale image parsing determines the optimal size of each observation by an artificial agent at a given point in time while searching for the anatomical landmark. The artificial agent begins searching image data with a coarse field-of-view and iteratively decreases the field-of-view to locate the anatomical landmark. After searching at a coarse field-of view, the artificial agent increases resolution to a finer field-of-view to analyze context and appearance factors to converge on the anatomical landmark. The artificial agent determines applicable context and appearance factors at each effective scale.

Abstract: A method and system for virtual percutaneous valve implantation is disclosed. A patient-specific anatomical model of a heart valve is estimated based on 3D cardiac medical image data and an implant model representing a valve implant is virtually deployed into the patient-specific anatomical model of the heart valve. A library of implant models, each modeling geometrical properties of a corresponding valve implant, is maintained. The implant models maintained in the library are virtually deployed into the patient specific anatomical model of the heart valve to select an implant type and size and deployment location and orientation for percutaneous valve implantation.

Abstract: A method for generating a personalized scaffold for an individual includes acquiring images of an anatomy of interest corresponding to an intended scaffold location and acquiring test results related to the anatomy of interest. One or more functional specifications are generated based on the images and test results and one or more scaffold parameters are selected based on the functional specifications. A final scaffold may then be generated using the one or more scaffold parameters.

Abstract: Systems and methods are provided for automatic classification of multiple abnormalities that are visible in chest X-ray images. The systems and methods are based on a deep learning architecture that predicts, in addition to classification scores of abnormalities, lung/heart masks, and the location of certain abnormalities. By training a multi-task network to improve all the results, the network and the resulting abnormality classification is improved. Normalization of the chest X-ray images is also used to improve the accuracy and efficiency of the multi-task network.

Abstract: Systems and methods for image classification include receiving imaging data of in-vivo or excised tissue of a patient during a surgical procedure. Local image features are extracted from the imaging data. A vocabulary histogram for the imaging data is computed based on the extracted local image features. A classification of the in-vivo or excised tissue of the patient in the imaging data is determined based on the vocabulary histogram using a trained classifier, which is trained based on a set of sample images with confirmed tissue types.

Abstract: For processing a medical image, medical image data representing a medical image of at least a portion of a vertebral column is received. The medical image data is processed to determine a plurality of positions within the image. Each of the plurality of positions corresponds to a position relating to a vertebral bone within the vertebral column. Data representing the plurality of positions is processed to determine a degree of deformity of at least one vertebral bone within the vertebral column.

Abstract: Systems and method are described for determining a malignancy of a nodule. A medical image of a nodule of a patient is received. A patch surrounding the nodule is identified in the medical image. A malignancy of the nodule in the patch is predicted using a trained deep image-to-image network.