Intelligent Assisted Diagnosis and Treatment System for Vascular Infarction Based on YOLO-SAM and Unique3D

Cardiovascular and cerebrovascular diseases have become the leading cause of death among chronic diseases in China. To address this, an intelligent vascular plaque detection and thrombosis risk prediction system was designed, integrating image recognition and 3D modeling into a multi-stage diagnostic framework. First, the YOLO model performs real-time plaque detection within vascular images, feeding the detected regions as input to the SAM model for precise segmentation. Subsequently, the segmented regions undergo high-fidelity 3D reconstruction using Unique3D technology to restore the spatial structure of vessels and plaques. Finally, an LSTM-based machine learning algorithm is trained using the 3D morphology and clinical labels to predict the probability of plaque-induced vascular infarction, thereby completing the closed-loop auxiliary diagnosis system.