A hippocampus segmentation method, device and electronic equipment applied to MRI

Abstract

The embodiment of the invention provides a hippocampal body segmentation method, a device and an electronic device applied to MRI. The method comprises the following steps: acquiring MRI three-dimensional images; constructing a target image having a preset size and containing a sensitive region of the MRI three-dimensional image; inputting the target image into at least two interactive neural networks trained in advance respectively, and obtaining the pre-segmentation results output by each interactive neural network; determining a segmentation result of the MRI three-dimensional image according to a predetermined weight of each of the interactive neural networks and a pre-segmentation result output by each of the interactive neural networks. The embodiment of the invention can improve theefficiency of segmenting the hippocampus.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a hippocampus segmentation method, device and electronic equipment applied to MRI. Background technique [0002] The hippocampus is an important part of the human brain nervous system, mainly responsible for memory and learning. The short-term memory in daily life is stored in the hippocampus, so the hippocampus plays a vital role in human memory and sense of direction. Changes in volume and shape of the hippocampus are associated with many neurological diseases. At present, volume atrophy or changes in shape of the hippocampus have been proven to be a prominent feature of mild cognitive impairment and Alzheimer's disease. Therefore, the morphology of the hippocampus Clinical analysis is of great significance in clinical practice. [0003] Magnetic Resonance Imaging (MRI) plays an important role in medical imaging technology due to its non-invasiveness, high resolution an...

AI Technical Summary

Problems solved by technology

Among them, the manual segmentation method is based on the doctor's experience in segmenting the hippocampus in the brain MRI. However, the manual segmentation method is time-consuming, inefficient and highly subjective, and it is difficult to achieve a good balance between accuracy and time-consuming

The atlas registration segmentation method is based on matching single or multiple atlases with the target image to obtain the segmentation results for the target image. However, the atlas registration method requires a lot of prior knowledge, and it is also difficult to achieve a good balance between accuracy and time consumption. the balance

Images