Methods, Computer Programs, and Devices for Defect Detection Artificial Intelligence Pipeline Computing Systems (Defect Detection Artificial Intelligence Pipeline Computing Systems)

摘要

PROBLEM TO BE SOLVED: To provide a lesion detection and classification artificial intelligence (AI) pipeline including a plurality of trained machine learning (ML) computer models. A first ML model processes an input volume (VOI) of a medical image to determine if the VOI exhibits a predetermined amount of anatomy. The AI pipeline determines if criteria such as a predetermined amount of anatomy of interest being shown on the input volume are met by the output of the first ML model. When satisfied, a second ML model processes the VOI to detect lesions corresponding to the subject's anatomy, and a third ML model is associated with lesion segmentation and the same lesion. A lesion treatment operation is performed, including performing lesion contour coupling and a fourth ML model processing a list of lesions to classify lesions. The AI pipeline outputs a list and classification of lesions. [Selection diagram] Fig. 1