Computed tomography medical imaging stroke model

US11331056B2 · kind B2 · utility

Assignees

• GE Precision Healthcare LLC · US
• Health Care Systems, Inc. · US
• The General Hospital Corporation · US
• The Brigham and Women's Hospital, Inc. · US

Inventors

• John F. Kalafut · Pittsburgh, US
• Bernardo Bizzo · Boston, US
• Romane Isabelle Marie-Bernard Gauriau · Paris, FR
• Michael Lev · Newton, US
• Mark Michalski · Menomonee Falls, US

Key dates

• Filing date: Sep 30, 2019
• Grant date: May 17, 2022
• Expiry date: May 26, 2040

Classification

• Technology area (CPC G): Physics
• CPC primary: G06T2207/30016
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Systems and techniques for generating and/or employing a computed tomography (CT) medical imaging stroke model are presented. In one example, a system employs a convolutional neural network to generate learned medical imaging stroke data regarding a brain anatomical region based on CT data associated with the brain anatomical region and diffusion-weighted imaging (DWI) data associated with one or more segmentation masks for the brain anatomical region. The system also detects presence or absence of a medical stroke condition in a CT image based on the learned medical imaging stroke data.