Decision support system for medical therapy planning

Abstract

A decision support system for medical therapy planning is provided. For decision support in a medical therapy, machine learning provides a machine-learned generator for generating a prediction of outcome for therapy personalized to a patient. Deep learning may result in features more predictive of outcome than handcrafted features. More comprehensive learning may be provided by using multi-task learning, where one of the tasks (e.g., segmentation, non-image data, and/or feature extraction) is unsupervised and/or draws on a greater number of training samples than available for outcome prediction alone.

Description

technical field [0001] This patent document requires the filing of Provisional U.S. Patent Application Serial No. 62 / 677,716, filed May 30, 2018, and Provisional U.S. Patent Application Serial No. 62 / 745,712, filed October 15, 2018, under 35 U.S.C. § 119(e) Said Provisional United States Patent Application is hereby incorporated by reference for the benefit of the date hereof. Background technique [0002] This embodiment relates to decision support for therapy. A typical example is the application in radiation therapy. Radiation therapy is a useful and cost-effective treatment strategy for many types of cancer. Although radiation therapy is an effective cancer treatment, most patients subsequently experience radio-resistance and recurrence of their cancer. Physicians seek to select treatment based on the specific characteristics of the patient and their disease in order to avoid treatment resistance and relapse. [0003] Predictors of radiation response were largely lim...

AI Technical Summary

Problems solved by technology

Radiological analysis may not maximize the information obtained, where a large number of features are often extracted from images containing a large amount of redundant or i

Images