Craniocerebral CT image bleeding area automatic segmentation method

Abstract

The invention discloses a craniocerebral CT image bleeding area automatic segmentation method. The method comprises the steps of: S1, obtaining a craniocerebral CT image needing bleeding area automatic segmentation; and S2, enabling the craniocerebral CT image to sequentially pass through a pre-constructed and trained improved U-Net convolutional neural network, wherein the overall structure of the network is three down-sampling layers and three up-sampling layers, and in the jump connection process, feature maps obtained through down-sampling are copied, convolved and cut respectively and then spliced with feature maps behind the corresponding up-sampling layers, and the corresponding up-sampling layers carry out up-sampling and convolution on the spliced feature maps. The improved U-Netconvolutional neural network totally comprises seven layers, so that information loss is reduced while feature extraction is ensured, time is saved, and the overall efficiency is improved; and in thejump connection step, convolution operation is added to a down-sampled feature map, more information can be provided for an up-sampling layer while the number of model layers is not changed, and therefore subsequent image segmentation of the cerebral hemorrhage CT image is improved.

Description

technical field [0001] The invention belongs to the field of medical image processing, and in particular relates to a method for automatically segmenting hemorrhage areas in cranial CT images. Background technique [0002] According to the Global Burden of Disease, stroke is the second most common cause of death and one of the leading causes of disability worldwide. A stroke occurs when blood flow to a specific area of ​​the brain is interrupted. When this happens, brain cells are deprived of oxygen and die. When a stroke causes brain cells to die, functions controlled by this area of ​​the brain, such as memory and muscle control, are lost. A hemorrhagic stroke occurs when a blood vessel ruptures, while an ischemic stroke occurs when a blood vessel becomes blocked, causing more damage than an ischemic stroke. [0003] In the medical field, in order to meet the needs of disease diagnosis and treatment plan formulation, it is often necessary to scan patients to determine th...

AI Technical Summary

Problems solved by technology

[0004] The technical purpose of the present invention is to provide a method for automatic segmentation of hemorrhage areas in cranial CT images to solve the technical problems of slow image recognition speed and low precision

Images