Development Of White Matter Fiber Bundle-based Computing Platform And Its Application In Migraine Research

Fiber bundle tracking technology based on diffusion tensor imaging data can non-invasively detect white matter fiber bundles in the living brain.By analyzing the white matter information in specific areas,we can understand the local microstructural changes of the fiber bundle,which helps people study the effects of some neurological diseases on the white matter of the brain.At present,the international mainstream white matter analysis methods mainly include Tract-based spatial statistics(TBSS)and Automating Fiber-Tract Quantification(AFQ).This method of TBSS adopts the idea of "group average skeleton",obtains the "average white matter skeleton map" of the subjects based on the FA image,and projects individual FA parameters onto their respective FA white matter skeletons,so as to achieve the group level of different subjects’ white matter Accurate statistical analysis.TBSS reduces the sensitivity of analysis results to registration accuracy by introducing skeletonization and projection steps.However,the TBSS method can only analyze the white matter fiber bundles at the skeleton,but not the white matter structure in other areas.In 2012,Jason D.Yeatman and others from Stanford University proposed the AFQ method,and developed the AFQ open source toolkit based on Matlab.Users can use the 18 white matter fiber bundles built into the automated analysis software of the toolkit.However,the internal corresponding method of the fiber bundle used in AFQ is not accurate enough,and because the AFQ method is based on the characteristics of individual fiber bundle tracking,the use of this method is greatly affected by the quality of individual data.Our team has proposed a method called Tractography atlas-based analysis(TABS),which can parametrically model white matter fiber bundles to accurately locate white matter microstructure damage.Based on this,this paper improves the method in view of the current registration errors and the problem of insufficient fiber bundle authenticity.By combining probabilistic and deterministic fiber bundle tracking methods,the extracted individual white matter microstructure features are screened to improve the accuracy of the TABS analysis results.In addition,this article integrates the improved TABS method into a white matter analysis platform to provide convenience for other researchers to use the TABS method.Migraine is a common primary headache disease,which is more common among women.In addition,there are significant differences in symptoms and treatment effects from men.At present,some studies have found that certain gender-related migraine patients have certain disease-related differences in brain structure and function.This paper analyzes the white matter fiber bundles of migraine patients based on the improved TABS method.The results show that the improved steps in this paper can filter out the noise characteristics.At the same time,we found that the fiber bundles from the thalamus to the forebrain island,the dorsal buckle,and the primary somatosensory cortex are complete Sex has an interactive effect between gender and migraine disease,which may be a factor in gender differences in all aspects of migraine.