Case Western Reserve University

Rutgers, The State University of New Jersey

University of Pa.

Satish E Viswanath

Anant Madabhushi

Pallavi Tiwari

Jacob Antunes

Zhouping Wei

Cheng Lu

Michael D Feldman

Mark Rosen

Robert Toth

Iulia Barbur

John Tomaszeweski

Charlems Alvarez Jimenez

Thomas Desilvio

Amir Reza Sadri

In some embodiments, the present disclosure relates to a non-transitory computer-readable medium storing computer-executable instructions that, when executed, cause a processor to perform operations, including forming an imaging dataset having imaging data corresponding to one or more radiological images of a patient having a bowel disease; operating upon the imaging data with one or more convolutional neural network (CNN) segments configured to generate a plurality of CNN outputs, the one or more CNN segments respectively including a convolution layer configured to perform a convolution on the imaging data; and applying a wavelet network to the plurality of CNN outputs to generate a plurality of convolution wavelet network (CWN) outputs, the wavelet network being configured to decompose the plurality of CNN outputs according to a mother wavelet. A predictive signature associated with disease response or risk is constructed using the plurality of CWN outputs.

Hybrid convolutional wavelet networks for predicting treatment response via radiological images of bowel disease

The present disclosure, in some embodiments, relates to a non-transitory computer-readable medium storing computer-executable instructions. The computer readable medium is configured to cause a processor to access an image volume of a rectum comprising a rectal tumor. A forward mapping is generated based on non-rigidly registering a healthy rectal atlas to the image volume. The forward mapping is inverted to generate an inverse mapping from the image volume to the healthy rectal atlas. Based on the inverse mapping, a plurality of deformation vectors, associated with a deformation within a rectal wall of the rectum, are determined. Magnitude based deformation features and orientation based deformation features are computed from the plurality of deformation vectors. One or more of the magnitude based deformation features and one or more of the orientation based deformation features are utilized to determine a response of a patient to a chemoradiation treatment.

Structural rectal atlas deformation features for characterizing intra-wall and peri-wall chemoradiation response on magnetic resonance imaging (MRI)

Embodiments discussed herein facilitate predicting response to therapy in Crohn's disease. A first set of embodiments discussed herein relates to accessing a radiological image of a region of tissue demonstrating Crohn's disease associated with a patient; defining a mesenteric fat region by segmenting mesenteric fat represented in the radiological image; extracting a set of radiomic features from the mesenteric fat region; providing the set of radiomic features to a machine learning classifier configured to compute a probability of response to therapy in Crohn's disease based, at least in part, on the set of radiomic features; receiving, from the machine learning classifier, a probability that the region of tissue will respond to therapy; generating a classification of the patient as a responder or non-responder based, at least in part, on the probability; and displaying the classification.

Predicting response to therapy for adult and pediatric crohn's disease using radiomic features of mesenteric fat regions on baseline magnetic resonance enterography

Embodiments discussed herein facilitate determination of responsiveness to chemoradiation treatment in rectal cancer patients based on structural deformation features obtained from a pre- or post-treatment medical imaging. One example embodiment can perform operations comprising: accessing an image volume of a rectum comprising a rectal tumor; generating a forward mapping based on non-rigidly registering a healthy rectal atlas to the image volume; inverting the forward mapping to generate an inverse mapping from the image volume to the healthy rectal atlas; determining, based on the inverse mapping, an associated deformation magnitude for each voxel of a plurality of voxels associated with the rectum; computing one or more structural deformation features based on the associated deformation magnitudes for the plurality of voxels; and predicting via a classifier whether or not the rectal tumor will respond to the chemoradiation treatment based at least in part on the one or more structural deformation features.

Methods and apparatus associated with predicting colorectal cancer tumor invasiveness are described. One example apparatus includes a set of circuits, and a data store that stores radiological images of tissue demonstrating colorectal cancer. The set of circuits includes a circumferential resection margin (CRM) prediction circuit that generates a CRM probability score for a diagnostic radiological image, an image acquisition circuit that acquires a diagnostic radiological image of a region of tissue demonstrating colorectal cancer pathology and that provides the diagnostic radiological image to the CRM prediction circuit, and a training circuit that trains the CRM prediction circuit to quantify chemoradiation response in the region of tissue represented in the diagnostic radiological image. The training circuit trains the CRM prediction circuit using a set of composite images.

Treatment planning and evaluation for rectal cancer via image analytics

Embodiments access an image of a region of interest (ROI) demonstrating cancerous pathology; extract radiomic features from the ROI; define a radiomic feature expression scene based on the ROI and radiomic features; generate a cluster map by superpixel clustering the expression scene; generate an expression map by repartitioning the cluster map into expression levels; compute a textural and spatial phenotypes for the expression map based on the expression levels; construct a radiomic spatial textural (RADISTAT) descriptor by concatenating the textural and spatial phenotypes; provide the RADISTAT descriptor to a machine learning classifier; receive, from the machine learning classifier, a first probability that the ROI is a responder or non-responder, or a second probability that the ROI will experience long-term survival or short-term survival, based, at least in part, on the RADISTAT descriptor; and generate a classification of the ROI as a responder or non-responder, or long-term survivor or short-term survivor.

Characterizing intra-tumoral heterogeneity for response and outcome prediction using radiomic spatial textural descriptor (RADISTAT)

Methods, apparatus, and other embodiments associated with predicting Crohn's Disease (CD) patient response to immunosuppressive (IS) therapy using radiomic features extracted from diagnostic magnetic resonance enterography (MRE). One example apparatus includes an image acquisition circuit that acquires an MRE image of a region of tissue demonstrating CD pathology, a segmentation circuit that segments a region of interest (ROI) from the diagnostic radiological image, a classification circuit that extracts a set of discriminative features from the ROI and that distinguishes the ROI as a responder or non-responder to IS therapy, and a CD prediction circuit that generates a radiomic enterographic (RET) score based on the diagnostic radiological image or the set of discriminative features. A prognosis or treatment plan may be provided based on the RET score.

Radiomic features on diagnostic magnetic resonance enterography

The present invention provides a system and method for analysis of multimodal imaging and non-imaging biomedical data, using a multi-parametric data representation and integration framework. The present invention makes use of (1) dimensionality reduction to account for differing dimensionalities and scale in multimodal biomedical data, and (2) a supervised ensemble of embeddings to accurately capture maximum available class information from the data.

Enhanced multi-protocol analysis via intelligent supervised embedding (empravise) for multimodal data fusion

This invention relates to computer-assisted diagnostics and classification of prostate cancer. Specifically, the invention relates to segmentation of the prostate boundary on MRI images, cancer detection using multimodal multi-protocol MR data; and their integration for a computer-aided diagnosis and classification system for prostate cancer.

Growing community of inventors

Pepper Pike, OH, United States of America

Satish E Viswanath