Pancreatic cancer accurate diagnosis system based on PET/CT double-time imaging

Abstract

The invention discloses a pancreatic cancer accurate diagnosis system based on PET / CT double-time imaging, which comprises: a dual-time image registration module is used for registering PET / CT delayedscanning image data acquired through PET / CT image scanning with PET / CT early scanning image data; a focus segmentation module which is used for carrying out interactive semi-automatic segmentation onthe early PET image by adopting a deep interactive segmentation network model based on a convolutional neural network; and an imaging omics classification diagnosis module which is used for analyzingthe segmented PET / CT early scanning image and PET / CT delayed scanning image to obtain a diagnosis result. According to the method, the segmentation error caused by the partial volume effect of the PET image is reduced, the segmentation habit of a current clinician is automatically learned, the precision and robustness of automatic segmentation of a single CNN network are further improved, the false positive rate of pancreatic cancer diagnosis can be effectively reduced, and the system has good clinical popularization and application prospects.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a system for precise diagnosis of pancreatic cancer based on PET / CT dual-time imaging. Background technique [0002] In recent years, with the continuous improvement of people's living standards and medical diagnosis technology, the incidence of pancreatic diseases in my country and even in the world has a rising trend. Among them, the accurate identification between autoimmune pancreatitis (AIP) and pancreatic ductal adenocarcinoma (Pancreatic ductal adenocarcinoma, PDAC) has become a major clinical problem. Traditional imaging examination methods, such as ultrasound, computed tomography (CT) and magnetic resonance imaging (MRI), can diagnose patients with typical manifestations (sausage-shaped diffuse hyperplasia of the pancreas) by using the anatomical information of the pancreas. large) AIP patients. Positron emission tomography (PET) / computed tomography (C...

AI Technical Summary

Problems solved by technology

[0003] However, PET / CT mainly relies on fluorodeoxyglucose ( 18 F-FDG) can be used to judge good and evil by the level of metabolism of pancreatic lesions, which lacks specificity.

Images