Tumor recurrence prediction method, system, equipment and medium

• Main Author: MO ZIJIE
• Format: Patent
• Language: Chinese; English
• Published: 07.06.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATIONTECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING ORPROCESSING OF MEDICAL OR HEALTHCARE DATA; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The invention discloses a tumor recurrence prediction method, system and device and a medium. The prediction method comprises the following steps: acquiring a medical image of a patient before treatment; extracting image features from the medical image; and inputting the medical image and the image features into a tumor recurrence prediction model, predicting whether the tumor target region relapses or not, and segmenting the recurrence region. According to the method, medical images and fusion features before treatment are analyzed on the basis of deep learning, the model pays more attention to an image region with specific features, and the accuracy of tumor recurrence region segmentation is improved; the image features of the medical image are input into the auxiliary network model, and then the output of the auxiliary network model is input into the main network model, so that the main network model is assisted to obtain a better prediction effect; the multi-task network formed by the main network model a