Prompt learning pulmonary nodule segmentation method based on anchor points

The invention relates to a pulmonary nodule segmentation method based on an anchor point prompt learning encoder network, and belongs to the field of medical images and deep learning. The invention provides a new method for solving the problem that the current pulmonary nodule segmentation accuracy is insufficient. According to the method, a traditional Yolov5 model is improved, a prompt learning module is added in a Head module, a YoSAM network is constructed, and semi-automatic supervised learning is realized through anchor point feature extraction. According to the technical scheme, the method comprises the following steps that firstly, data processing is conducted on a disclosed chest medical image, and a two-dimensional image is manufactured; and secondly, performing data enhancement on the image, ensuring the balance of training and testing, transmitting a processed training set into a YoSAM network for training, storing an optimal model, obtaining a segmentation result, and performing comparison verifi