Pancreatic cyst recognition method and system based on semi-supervised learning

Abstract

The invention provides a pancreatic cyst recognition method and system based on semi-supervised learning, and solves the technical problem of difficulty in pancreatic cyst recognition in the prior art. The method comprises the following steps: training a semi-supervised learning model through a training set with a small number of artificial tags to obtain a trained semi-supervised learning model; extracting a pancreas contour probability graph of the multi-phase pancreas image by using the trained semi-supervised learning model; extracting a pancreatic cyst contour probability graph of the multi-phase pancreatic cyst image under the guidance of the pancreatic contour probability graph by using the trained semi-supervised learning model; and extracting a pancreatic cyst area of the pancreatic cyst contour probability graph, and recognizing benign and malignant pancreatic cyst according to the pancreatic cyst area and in combination with clinical knowledge driving features. According to the method, the pancreas and the pancreatic cyst are segmented step by step, and the pancreas contour probability graph obtained by pancreas segmentation is used for guiding the segmentation of the pancreatic cyst, so that the pancreatic cyst is segmented more accurately, and the accuracy of pancreatic cyst recognition is improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method and system for identifying pancreatic cysts based on semi-supervised learning. Background technique [0002] Driven by deep learning algorithms, the field of computer-aided diagnosis (CAD) has developed rapidly, especially in the field of medical image segmentation. However, due to the limitations of existing CT imaging, single-phase CT scans are often difficult to accurately locate the contours of organs, and the determination of the contours of lesions is more complicated. Since different phases can enhance different details, referring to different phases is an effective strategy to identify the boundaries of organs or lesions as completely as possible. In recent years, omics features driven by clinical knowledge can be well combined with features extracted by deep learning to further improve the performance of disease diagnosis and have been widely used in ma...

AI Technical Summary

Problems solved by technology

[0003] In view of the above problems, the embodiment of the present invention provides a method and system for identifying pancreatic cysts based on semi-supervised learning, which solves the technical problem of difficult identification of pancreatic cysts in the prior art

Images