Based on artificial intelligence technology, the prediction model is built by outlining the quantitative mapping correlation between annotated prostate cancer Whole Slide Images and MRI, and clarifying the common features. Firstly, the model can accurately diagnose the radical pathology of prostate cancer, which can be exempted from immunohistochemistry to obtain detailed pathological information; secondly, the established AI prediction model can accurately diagnose the benign/malignant, invasiveness, grade and subtype of prostate cancer by predicting the participant's MRI images before surgery or puncture, so that a personalised treatment plan can be formulated for the patient before operation or puncture. Finally, based on AI technology, the model learns from the MRI images and performs 3D reconstruction of the prostate and lesions before surgery/puncture, thus clarifying the exact location of the lesions and guiding puncture or surgical treatment.
