Cerebral ischemia auxiliary analysis method and system based on medical image processing

Abstract

The invention relates to the technical field of medical image processing, in particular to a cerebral ischemia auxiliary analysis method and system based on medical image processing. The method comprises the following steps: respectively inputting a brain CT image into an image classification network and a brain area detection network to obtain a first confidence coefficient and a first central axis; traversing the first central axis according to a preset step length, dividing the traversed initial symmetric line into comparison areas, calculating the cosine similarity of each group of comparison areas, and adjusting the final position to obtain an optimized final point so as to obtain a second central axis; re-dividing a plurality of groups of symmetric regions according to the second central axis, and symmetrically assigning values to each group of symmetric regions and inputting the values into the network to obtain a second confidence coefficient; and forming a sample sequence according to the first confidence coefficient and the second confidence coefficient, inputting the sample sequence into the deep neural network, and outputting a classification result about whether the symmetric region has a lesion region. The cerebral ischemia condition is effectively detected and positioned, and the positioning precision and the evaluation accuracy of cerebral ischemia lesion auxiliary analysis are improved.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a cerebral ischemia auxiliary analysis method and system based on medical image processing. Background technique [0002] Vascular disease, also known as stroke, is a neurological disease caused by various causes that cause the blood vessels supplying blood to the brain to develop lesions. It mainly includes ischemic and hemorrhagic cerebrovascular disease; of which ischemic cerebrovascular disease (also known as ischemic stroke) accounts for about 80%. Ischemic cerebrovascular disease refers to the degeneration, necrosis or transient loss of function of local brain tissue including nerve cells, glial cells and connecting fibers due to blood supply disorders. Cerebral artery occlusion caused by intravascular thrombosis, embolism, and vascular stenosis is the main cause of ischemic stroke. [0003] The existing methods for auxiliary analysis of cerebral ischemia...

AI Technical Summary

Problems solved by technology

[0003] The existing methods for auxiliary analysis of cerebral ischemia based on medical image processing usually use neural networks combined with network optimization methods such as self-attention mechanisms to achieve the purpose of auxiliary analysis. There are many kinds of methods. If you only rely on the results of the neural network output for auxiliary analysis, the results often have large errors, and it is difficult to really play the role of assisting doctors in analysis.

Images