Abstract: One aspect of the invention provides a computer-implemented method of identifying one or more clinical factors associated with an artificial intelligence prediction. The computer implemented method includes: applying a previously trained deep neural network to one or more images for a subject to produce a prediction, the previously trained deep neural network comprising a plurality of nodes; comparing outputs of the nodes to previously identified patterns of node outputs for a plurality of individual clinical factors; and identifying which of the plurality of individual clinical factors are most correlated with the outputs of the nodes.

Abstract: The described implementations relate to systems, methods, and apparatuses for generating regions of interest (214) from imaging data (212). Specifically, the regions of interest are generated for tracking treatment efficacy in a more consistent and repeatable manner. The regions of interest can be generated from contrast medium and non-contrast medium enhanced scans (102) of a patient. Voxel data derived from the scans can be collected and distributed according to respective intensity values in order to identify mode voxels (116, 118, 120) for particular ranges (128) of intensities. Regions of interest (110, 112, 114) can then be generated for each identified mode voxel, and standard deviations for the regions of interest can be determined. One or more thresholds can be derived from the determined standard deviations in order to further filter the intensity values and identify filtered groups of voxels to be the resulting regions of interest.

Abstract: The described implementations relate to systems, methods, and apparatuses for generating regions of interest (214) from imaging data (212). Specifically, the regions of interest are generated for tracking treatment efficacy in a more consistent and repeatable manner. The regions of interest can be generated from contrast medium and non-contrast medium enhanced scans (102) of a patient. Voxel data derived from the scans can be collected and distributed according to respective intensity values in order to identify mode voxels (116, 118, 120) for particular ranges (128) of intensities. Regions of interest (110, 112, 114) can then be generated for each identified mode voxel, and standard deviations for the regions of interest can be determined. One or more thresholds can be derived from the determined standard deviations in order to further filter the intensity values and identify filtered groups of voxels to be the resulting regions of interest.

Abstract: A system for transcatheter arterial chemoembolization (TACE) includes a visualization software module (115) configured to assess vascular geometry of an organ in an image of the organ. A tumor viability software module (124) is configured to provide a tumor viability map of the organ to be overlaid on the image of the organ. An imaging modality (126) is configured to track an instrument in or in proximity of the organ to ensure that the instrument is positioned within the organ for treatment in accordance with the tumor viability map.

Abstract: A system (100) and method which assess the results of uterine fibroid embolization. A fibroid lesion is three-dimensionally segmented by a segmentation module (110) on contrast-enhanced arterial phase MRI or CBCT images and a differentiation process performed by a subtraction module (114) identifies actual contrast enhancement from background enhancement. The three-dimensional segmentation mask is then applied to the differentiated image and a region of interest is defined by a comparison module (116) in order to define a normalized threshold. A computation module (112) performs a voxel-by-voxel analysis of enhancing fibroid volume and quantifies the viable enhanced fibroid volume and overall fibroid volume in absolute numerical figures.