Method and system for detecting postoperative prognosis data of pancreatic cancer patient

Abstract

The invention discloses a pancreatic cancer patient postoperative prognosis data detection method and system, and the method comprises the steps: constructing a 3D convolutional neural network model which is used for recognizing the CT imaging features of a target tumor of a pancreatic cancer patient; preoperative enhanced CT of a pancreatic cancer patient is collected, the preoperative enhanced CT is recognized through the 3D convolutional neural network model, and CT imaging features are generated; preoperative basic information and clinical examination data of a pancreatic cancer patient are collected, data processing is carried out through a hidden layer, and preoperative feature data of the pancreatic cancer patient are generated; based on the CT imaging features and the preoperative feature data, through the hidden layer, generating evaluation data for postoperative prognosis evaluation of the pancreatic cancer patient; according to the method, the economical efficiency and generalizability of the method are ensured, and compared with the prior art, the generated data improve the accuracy of prognosis evaluation.

Description

technical field [0001] The invention belongs to the field of medical artificial intelligence, and in particular relates to a detection method and system for postoperative prognosis data of pancreatic cancer patients. Background technique [0002] At present, the risk stratification and treatment decision evaluation of pancreatic cancer surgical prognosis still rely on traditional clinical TNM staging, but standard clinicopathological risk factors cannot accurately predict individual prognosis, and have limited role in guiding surgical risk-benefit assessment. In recent years, with the development of artificial intelligence in the medical field, the application of machine learning and its developed deep learning methods in various diseases has increased rapidly. Compared with traditional radiomics, which requires artificial determination of image features, deep learning, as an emerging technology in the field of artificial intelligence, can directly process the raw data of sa...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the technical problem of inaccurate detection of evaluation data in the existing pancreatic cancer prognosis evaluation, resulting in limited preoperative guidance and postoperative prognosis evaluation and prediction ability, the purpose of the present invention is to propose a method for postoperative prognosis data of pancreatic cancer patients. The detection method and system can solve the problem of inaccurate detection of prognosis data of pancreatic cancer patients after operation, and improve the accuracy of basic curative effect evaluation before operation.

Images