Method for identifying cerebral cortex surface marking region of three-dimensional cerebral magnetic resonance image

Abstract

The invention relates to a method for identifying a cerebral cortex surface marking region of a three-dimensional cerebral magnetic resonance image. The method is characterized by comprising the following steps of: defining a structural connection mode of a cerebral cortex surface and a subcutaneous tissue layer by using structural connection information of the cortex surface and subcutaneous tissue layer; classifying the connection mode to realize division of the cerebral cortex surface region of a single body; and then extracting the stable and high-consistency cerebral cortex surface region between the bodies to serve as the cerebral cortex surface marking region. Compared with other methods, the cerebral cortex surface marking region is identified by using the stable and high-consistency connection mode of the cerebral cortex surface and the subcutaneous tissue layer between the bodies, and the obtained cerebral cortex surface marking region is more stable and consistent.

Description

technical field [0001] The invention relates to a method for identifying marker regions on the surface of the cerebral cortex in a three-dimensional brain magnetic resonance image, which uses the nerve fiber connection pattern between the surface of the cerebral cortex and the subcortical tissue in the three-dimensional brain magnetic resonance image to identify the marker region on the surface of the cerebral cortex. It belongs to the field of medical image processing, computational neuroanatomy and other fields. Background technique [0002] In three-dimensional brain magnetic resonance images, the surface of the human cerebral cortex is mainly composed of sulci and gyri, and its anatomical structure is extremely complex. The anatomical geometric patterns of the sulci and gyri on the surface of the cerebral cortex vary greatly between different people. In the field of brain imaging and neuroscience, it is generally believed that in different individuals, the cortical surfa...

AI Technical Summary

Problems solved by technology

[0005] The main disadvantages of the above four methods for identifying landmarks on the surface of the cerebral cortex are that the first method of expert marking is limited by personal manpower and knowledge, is too subjective, has poor repeatability, and is not suitable for processing large-scale data sets; The two clustering methods largely depend on several key clustering parameters in the algorithm. The clustering results obtained by different clustering parameters are different. The parameter changes, and the parameters need to be adjusted according to different data sets during the clustering process; the third map-based method is based on the premise that the expert provides the landmark map, and the image registration is the basis. If the brain to be marked and the template brain When the anatomical structure is too different, the result is limited by the registration algorithm, and the accuracy is greatly reduced; although the fourth method can accurately obtain the landmark area of ​​the cerebral cortex, it is necessary to design an MRI scan task for each individual brain for each acquisition. Activate and scan the brain for data collection, the activation results are very sensitive to the parameters set during the scanning process, and the functional areas activated by different tasks are limited, the obtained landmark areas are limited and the experimental cost is high, in addition, due to the long scanning time, the process is complicated , the scanning task is limited by conditions such as the physical health of the experimental subjects, for example, it is difficult to obtain the landmark areas of the elderly, children and other objects by designing functional imaging tasks

Images