Brain tumor segmentation method based on multi-level structure relation learning network

Abstract

The invention provides an advanced multi-level structure relationship learning network for segmenting brain tumor data. In each subnet, an environmental information mining module is introduced betweenan encoder and a decoder, and environmental information in a single domain and environmental information between different domains are respectively mined by adopting a dual self-attention mechanism and spatial interaction learning.

Description

technical field [0001] The invention relates to a tumor segmentation technology, in particular to a brain tumor segmentation method based on a multi-level structure relational learning network. Background technique [0002] Segmentation of brain tumors, segmenting different types of tumor regions in multimodal 3D magnetic resonance images. Brain tumor segmentation based on MRI data is an important academic and industrial topic and has been an active research area in the past decade. Efficient and fast brain tumor segmentation is helpful for neurological status monitoring, assessment of tumor development, and diagnosis of encephalopathy. [0003] In recent years, deep learning-based cascaded multi-layer networks and multi-scale analysis have made great progress in medical image segmentation. However, how to accurately classify each pixel is still a challenge in brain tumor segmentation based on MRI data. MRI data images have low contrast and different types of tumors have ...

AI Technical Summary

Problems solved by technology

[0005] For the second problem, the independent analysis and linear post-fusion of different domains make the segmentation model easy to fall into the local optimal problem

Images