Online Ventricular Segmentation System Based on Machine Learning

With the continuous expansion of China's medical rigid demand and the imbalance between supply and demand caused by insufficient medical resources, this gap provides an entry point for the combination of the Internet and the medical industry. As a result, network medical care has gradually entered people's attention. Therefore, this paper designs a network-based ventricular segmentation system that will be trained. The ventricular segmentation model is placed in the cloud, and the user only needs to input the Cardiac Magnetic Resonance Image (CMRI) through the network terminal to obtain the detection result. The model introduces the Mask R-CNN algorithm based on deep learning into the research of nuclear magnetic image edge detection, trying to solve the problem of less image, difficult marking and low edge precision. Ventricular segmentation is performed through a deep learning network. Thereby it improves the accuracy of CMRI edge detection.