System and method for automatic assessment of cancer
OwnerKavya Venkata Kota Sai KOPPARAPUinventor
Life of this patent
Estimating the term…
Abstract
Glioblastoma multiforme (GBM) is the most aggressive type of brain cancer. It is critical to determine the most effective patient-specific treatment quickly. Exemplary embodiments use a data-driven approach to extracting brain tumor information from data obtain from Whole Slide Image that is uploaded through an interface. Exemplary embodiments generate the following information about a glioblastoma tumor from a brain biopsy slide using neural networks: annotated areas of relevant tissues, molecular subtype, and expression status of an important gene and include three steps: the segmentation of tumor features; prediction of molecular subtype; and prediction of gene methylation status from a WSI.
Claims21
A computer implemented method for characterization of a brain tumor comprising:
- receiving, on at least one processor, data from a subject with a brain tumor;
- evaluating, using the at least one processor, the data using a classifier which is an electronic representation of a classification system, each said classifier trained using an electronically stored set of training data vectors, each training data vector representing an individual human and data for the respective human, each training data vector further comprising a classification with respect to the characterization of the brain tumor in the respective human;
- and
- outputting, using the at least one processor, a classification of the sample from the human test subject concerning the characterization of the brain tumor in the subject based on the evaluating step.
A computer implemented method, comprising:
- receiving, by a computer processor, a scanned biopsy WSI of a human brain of a subject having a brain tumor wherein the computer comprises a cloud-based server and wherein the scanned biopsy WSI is received through a web-based application;
- preprocessing, by a computer, the scanned biopsy WSI;
- processing, by the computer, the preprocessed scanned biopsy WSI using one or more machine learning architectures;
- outputting, by the computer through the use of one or more trained machine learning architectures, an annotated image, wherein the annotated image comprises tumor feature segmentation, a subtype of the tumor, and a MGMT promotor methylation status;
- and
- giving a capability, through the annotated image, to determine a personalized treatment plan for the subject with glioblastoma multiforme (GBM) based on the tumor's subtype and MGMT promotor methylation status.
A system, comprising:
- a server comprising at least one processor and memory comprising computer-readable instructions which when executed by the processor cause the processor to perform the steps comprising:
- receiving a scanned biopsy whole side image (WSI) of a human brain of a subject having a brain tumor comprising a Glioblastoma Multiforme (GBM) from a computer terminal that is located remotely from the server;
- preprocessing the scanned biopsy WSI by applying a tile-based image analysis framework to the scanned biopsy WSI;
- processing the preprocessed scanned biopsy WSI using three trained neural networks compromising one trained neural network for tumor features, one trained neural network for tumor subtype, and one trained neural network for MGMT promoter methylation status;
- and
- outputting an annotated image, wherein the annotated image comprises smaller patches, a subtype of the tumor, and a MGMT promotor methylation status at MGMT+ or MGMT−.
21-76. (canceled)
Dependencies are inferred from the claim text.