Method and system for computer-aided diagnosis with deep learning model

• Main Authors: MULLICK RAKESH, GUALTIERI JAMES WILLIAM, ARADHARI, CHANDAN, KUMAR, MALLAPPA
• Format: Patent
• Language: Chinese; English
• Published: 06.05.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATIONTECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING ORPROCESSING OF MEDICAL OR HEALTHCARE DATA; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

Various methods and systems for computer-aided diagnosis are provided. In one embodiment, a method includes: acquiring a medical image of a subject with an imaging system; generating a computer-aided diagnosis for the medical image using a radiologist model associated with a radiologist of the institution, the radiologist model comprising a deep neural network trained with a plurality of diagnoses provided by the radiologist; displaying the medical image and the computer-aided diagnosis to a radiologist via a display device; and selectively updating one or more of a radiologist model, an institution model associated with the institution, and a geographic model associated with a geographic area containing the institution based on the medical image. In this manner, a deep neural network model configured as a digital twin of a radiologist may provide assistance to the radiologist. 本发明提供了用于计算机辅助诊断的各种方法和系统。在一个实施方案中，一种方法包括：用成像系统获取受试者的医学图像；用与机构的放射科医生相关联的放射科医生模型为医学图像生成计算机辅助诊断，该放射科医生模型包括借助由放射科医生提供的多个诊断进行训练的深度神经网络；经由显示