Liver and tumor segmentation method and system based on multitask deep convolution network

Abstract

The invention provides a liver and tumor segmentation method and system based on multitask deep convolution network. The method comprises the steps of conducting data pre-processing and expanding; establishing the multitask deep convolution network; establishing a supervising layer having inter-task constraint; conducting network training; generating data test result. The invention is advantageousin that the multitask network includes two pathways, which are respectively used for realizing the identification and segmentation of liver and liver tumor; through sharing of the pathway part, the volume of the deep convolution network can be reduced, and the demand of model learning on the training data amount can be reduced; through connecting to a characteristic extracting module relevant torespective task and a corresponding output module after the sharing of the pathways, the multitask output can be realized; the aim of mutual constraint of the liver and liver tumor can be achieved bymeans of a supervising and learning module of multitask geometrical association information; liver and liver tumor can be accurately identified and segmented.

Description

technical field [0001] The invention relates to a liver and its tumor segmentation method and system based on a multi-task deep convolutional network, in particular to a method for realizing complex target segmentation in an image through multi-task learning and inter-task association constraints, belonging to the field of machine learning and medical image analysis , which can be applied to but not limited to medical image segmentation tasks, etc. Background technique [0002] Intelligent healthcare is a popular application direction in the field of artificial intelligence. If the machine can automatically identify and analyze medical images, it can help doctors achieve accurate and individualized diagnosis and treatment, and reduce the risk of diagnosis and treatment. Among them, the automatic identification and segmentation of target tissues in medical images is a key sub-problem. If the medical images given to doctors have eliminated redundant information or marked imp...

AI Technical Summary

Problems solved by technology

In addition, due to the difficulty of tumor segmentation, there is currently a lack of methods that can automatically segment liver tumors

[0003] At present, under-segmentation is prone to occur in the segmentation of liver tissue with lesions, that is, the tumor tissue cannot be effectively classified as a part of the liver; while tumor identification and segmentation alone can easily divide the lesion area of ​​other tissues into it

Images