Brain tissue automatic sketching method and sketching system for MRI head image, computing device and storage medium

Abstract

The invention discloses a brain tissue automatic sketching method and sketching system for an MRI head image, a computing device and a storage medium, and the method comprises the steps: obtaining a preset number of T1MRI brain images, and making a brain tissue segmentation label of each image; performing same preprocessing and 3D block processing on the MRI brain image and the brain tissue segmentation label to obtain respective 3D data blocks; inputting the 3D data blocks into the constructed semantic segmentation neural network for training until the model is stably converged, and obtaining an optimal brain tissue segmentation neural network model; inputting the to-be-segmented image subjected to the same preprocessing and 3D block processing into the trained brain tissue segmentation neural network model to obtain a brain tissue segmentation result; and performing post-processing and edge detection on the brain tissue segmentation result to obtain a brain tissue contour sketching result. According to the method, automatic segmentation of the human brain tissue can be realized, the speed and accuracy of brain tissue segmentation are improved, and the robustness and adaptability of brain tissue segmentation are enhanced.

Description

technical field [0001] The present invention relates to the technical field of image recognition, in particular to a brain tissue automatic delineation method for MRI head images, a delineation system, a computing device and a storage medium. Background technique [0002] Magnetic resonance imaging (MRI) allows humans to observe the brain tissue structure of the human brain non-destructively. The normal human brain mainly includes three components: gray matter, white matter, and cerebrospinal fluid. Quantitative calculation and comparison of these three components of the brain, and the premise of quantitative calculation is the scientific segmentation of these three brain components. [0003] At present, clinical doctors mainly rely on classical techniques such as threshold method, region-based method, clustering classification method, etc. for the segmentation of brain tissue. However, the above-mentioned methods are sensitive to noise, have instability, and face great chal...

AI Technical Summary

Problems solved by technology

[0003] At present, doctors mainly rely on classical techniques such as threshold method, region-based method, clustering classification method, etc. for the segmentation of brain tissue in clinical practice. However, the above methods are sensitive to noise, unstable, and face great challenges.

In the field of cognitive neuroscience, for complete brain imaging, brain tissue coordinate breakthrough systems such as those established by the Montreal Neurological Institute (MNI) based on a batch of MRI images of normal human brains are used to conduct brain imaging. Tissue segmentation method, but this method has a slow segmentation speed and requires slow and fine three-dimensional image registration, which has a large time cost and relies on the complete scan image of the human brain to perform

Images