基于动态Hurst指数的人脑老化复杂度分析

Complexity analysis of human brain aging using dynamic Hurst exponent

目的研究健康成人在不同年龄段的功能特点可以更好地理解人脑的老化过程,Hurst指数可以测量静息态功能磁共振成像时间序列的复杂度。既往对大脑复杂度随年龄变化的研究,还存在样本量小、未考虑复杂度的动态变化等局限性。因而有必要在大样本被试上通过刻画复杂度的动态变化特性,来研究人脑随年龄变化的规律。方法选择19~80岁的458名健康成人静息态功能磁共振成像数据,划分为青年组(n=213)、中年组(n=142)和老年组(n=103),通过滑窗法来计算动态Hurst指数,并进一步提取其平均值、标准差和变异系数特征来描述人脑的动态复杂度水平和波动程度,最后通过统计分析来评判不同年龄组被试的复杂度差异。结果在平均值特征上,双侧颞上回、双侧岛回、右侧扣带回和双侧腹中侧枕叶皮层的复杂度随年龄增长而降低,而双侧眶回和左侧楔前叶的复杂度则随年龄的增长而呈现出倒U形变化;在标准差和变异系数特征上,老年组在右侧楔前叶的动态复杂度显著大于青年组和中年组。结论人脑老化过程中全脑复杂度动态变化模式并不一致,且在不同的年龄段上,局部脑区存在不同的组间统计显著差异模式。该结果可以为研究者理解正常老化的功能机制提供帮助。

Objective It can better help to understand the aging process of human brain by studying the functional characteristics of healthy adults at different ages.Hurst exponent can measure the complexity of resting-state functional magnetic resonance imaging(rs-fMRI)time series.Previous studies on the change of brain complexity with age still have some limitations,such as small sample size and ignoring the dynamic change of complexity.Therefore,it is necessary to study the changing characteristics of the aging brain by measuring the dynamic changes of complexity in a large sample size.Methods By using the rs-fMRI data from 458 healthy adults aged 19 to 80,we firstly divided the subjects into the young group(n=213),the middle-aged group(n=142)and the elderly group(n=103).The dynamic Hurst exponent was then calculated by the sliding window method,and its mean value,standard deviation and coefficient of variation were further extracted to describe the level of dynamic complexity and fluctuation degree of the human brain.Finally,the complexity differences of groups with different ages were analyzed by statistical analysis.Results For the mean value,the complexity of bilateral superior temporal gyrus,bilateral insular gyrus,right cingulate gyrus and bilateral ventral middle occipital cortex could decrease with age,while the complexity of bilateral orbital gyrus and left precuneus showed an inverted U-shaped change with age.For the standard deviation and coefficient of variation,the dynamic complexity of the right precuneus in the elderly group was significantly higher than that in the young group and middle-aged group.Conclusions During the normal aging of the human brain,the dynamic change patterns of complexity in the whole brain are not consistent,and there are different statistically significant difference patterns among different age groups in certain local brain regions.Our results can help to improve the understanding of the functional mechanism of brain aging.