Segmentation method of cardiac structure in MRI images based on multi-channel convolutional neural network

Abstract

The present invention relates to a method for segmenting cardiac structures in MRI images based on a multi-channel convolutional neural network, which includes collecting heart movie MRI training data of normal people and heart patients, and manually marking the cardiac structures in the training data by experienced doctors as Heart segmentation labeling results, based on the training data to train the heart region extraction model, so that the heart region extraction model can accurately extract the heart region, and train the heart segmentation network according to the heart region extracted from the training data to segment the heart Structure, using the standard segmentation labeling results as a standard to measure the segmentation performance of the constructed heart segmentation network. The present invention uses a heart region extraction model based on a generative confrontation network to extract the heart, which improves the accuracy of the heart region extraction; at the same time, the context information between adjacent layers is used through a multi-channel convolutional neural network, which improves the segmentation accuracy and accuracy. Spend.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a method for segmenting cardiac structures in MRI images based on a multi-channel convolutional neural network. Background technique [0002] According to the statistics of the World Health Organization, cardiovascular disease is the disease with the highest mortality rate in the world, and about 19.7 million people died of cardiovascular disease in 2016. In clinical practice, cardiac function analysis plays an important role in heart disease diagnosis, risk assessment, patient management, and treatment decision-making. This is usually done with the aid of digital images of the heart to quantify global or regional cardiac function by assessing a range of clinical parameters such as ventricular volume, ejection fraction, stroke volume, myocardial mass, etc. Due to the good discrimination of soft tissues, the evaluation of left and right ventricular ejection fraction, strok...

AI Technical Summary

Problems solved by technology

The method based on deep learning can obtain more accurate automatic segmentation results. However, the existing heart segmentation methods based on deep learning mostly use 2D segmentation methods without considering the inter-layer context information. The inter-layer context information is important for accurate segmentation and improved segmentation. performance is valuable

Ignoring inter-layer contextual information is not in line with the actual workflow of clinicians

At the same time, due to the thick and large spacing of the scanning layers of the cardiac cine MRI image, directly using the inter-layer context information through the 3D segmentation method is not only computationally expensive but may not bring about performance improvement.

Images