Various embodiments include methods and diagnostic systems implementing the methods for determining a prognostic prediction of a likelihood of success or a likelihood of complications of an electrophysiology procedure at the identified area of electrophysiological interest. Various embodiments may include generating a patient-specific three-dimensional (3D) cardiac activation and arrythmia localization model identifying an area of electrophysiological interest for performing an electrophysiology procedure to treat the arrythmia, using the 3D heart model to identify heart structures near the identified area of electrophysiological interest, determining a prognostic indication of an electrophysiology procedure performed at the identified area of electrophysiological interest based at least in part on one or more heart structures near the area of electrophysiological interest, and generating an output providing a prognostic indication of an electrophysiology procedure at the identified area of electrophysiological interest based at least in part on the determined likelihood of success.

Methods and systems for training a model labeling two or more organic structures within an image. One method includes receiving a set of training images. The set of training images including a first plurality of images and a second plurality of images. Each of the first plurality of images including a label for each of the two or more organic structures and each of the second plurality of images including a label for only a subset of the two or more organic structures. The method further includes training the model using the first plurality of images, the second plurality of images, and a label merging function mapping a label from the first plurality of images to a label included in the second plurality of images.

A facility for processing a medical imaging image is described. The facility applies to the image a first machine learning model trained to recognize a view to which an image corresponds, and a second machine learning model trained to identify any of a set of anatomical features visualized in an image. The facility accesses a list of permitted anatomical features for images corresponding to the recognized view, and filters the identified anatomical features to exclude any not on the accessed list. The facility causes the accessed image to be displayed, overlaid with a visual indication of each of the filtered identified anatomical features.

Methods, systems, and apparatuses, including computer programs for detecting pancreatic neoplasms. A method includes providing an image as an input to a first model, obtaining first output data generated by the first model based on the first model's processing of the image, the first output data representing a portion of the image that depicts a pancreas, providing the first output data as an input to a second model, obtaining second output data generated by the second model based on the second model's processing of the second input data, the second output indicating whether the depicted pancreas is normal or abnormal, providing the first output data and the second output data as an input to a third model, and obtaining third output data generated by the third model, the third output data including data indicating that the pancreas is normal or data indicating a likely location of a pancreatic neoplasm.

A method of implementing an artificial intelligence based neuroradiology platform for neurological tumor identification comprises providing a multilayer convolutional network for neurological tumor identification configured for segmenting data sets of full neurologic scans into resolution voxels; supervised learning and validation of the platform by classification of tissue within classification voxels of a specific given training and validation data sets by the multilayer convolutional network for neurological tumor identification with each classification voxel of the training and validation data sets having a predetermined ground truth; and implementing the platform by classification of tissue within classification voxels of a specific given patient data sets by the multilayer convolutional network for neurological tumor identification with each classification voxel of each data set assigned a label. The platform may be used for T-cell therapy initiation and tracking. An artificial intelligence based neuroradiology platform implemented according to the method is disclosed.

Systems and techniques that facilitate an augmented inspector interface with targeted, context-driven algorithms are provided. In various embodiments, a magnification component can magnify a portion of a medical image. In various embodiments, a recognition component can recognize an anatomical structure depicted in the portion of the medical image. In various embodiments, a recommendation component can recommend one or more sets of computing algorithms or computing operations related to the anatomical structure. In various embodiments, a menu component can display the one or more recommended sets of computing algorithms or computing operations in a drop-down menu.

An apparatus of machine learning includes processing circuitry. The processing circuitry uses a first calibration model that receives, as input, first processing data and a first processing label assigned by a user to the first processing data and outputs calibration data relating to calibration of individual characteristics in label assignment by the first user, and trains a target model based on at least the first processing data and the calibration data or a calibrated label having individual characteristics calibrated using the calibration data.

A candidate generator generates a set of candidate three-dimensional image patches from an input volume. A candidate classifier classifies the set of candidate three-dimensional image patches as containing or not containing disease. Classifying the set of candidate three-dimensional image patches comprises generating an attention mask for each given candidate three-dimensional image patch within the set of candidate three-dimensional image patches to form a set of attention masks, applying the set of attention masks to the set of candidate three-dimensional image patches to form a set of masked image patches, and classifying the set of masked image patches as containing or not containing the disease. The candidate classifier applies soft attention and hard attention to the three-dimensional image patches such that distinctive image regions are highlighted proportionally to their contribution to classification while completely removing image regions that may cause confusion.

For robust view classification and measurement estimation in sequential ultrasound imaging, the classification and/or measurements for a given image or sequence of images are gated. To prevent oscillation in results, the gating provides consistent output.

Disclosed is a fully convolutional genetic neural network method for segmentation of infant brain record images. First, infant brain record image data is input and preprocessed, and genetic coding initialization is performed for parameters according to the length of a DMPGA-FCN network weight. Then, m individuals are randomly grouped into genetic native subpopulations and corresponding twin subpopulations are derived, where respective crossover probability and mutation probability pm of all the subpopulations are determined from disjoint intervals; and an optimal initialization value fa is searched for by using a genetic operator. Afterwards, fa is used as a forward propagation calculation parameter and a weighting operation is performed on the feature address featuremap. Finally, a pixel-by-pixel cross-entropy loss is calculated between predicted infant brain record images and standard segmented images to reversely update the weights, thus finally obtaining optimal weights of a network model for segmentation of the infant brain record images.