The Research Of Athletes’ Brain Plasticity Based On Windowed Dynamic Time Series Analysis

With the rapid development of brain science and artificial intelligence technology in recent years,scientists are eager to have a deeper understanding of the brain,and the study of brain plasticity has also become more and more popular.Athletes have been trained regularly for a long time and have good professional skills,which provides an ideal research sample for the study of brain plasticity.Currently,the understanding of the temporal dynamics of motor training on functional tissues of the brain is still limited.In this paper,we will explore the plasticity of the brain from time scale based on sliding window correlation method,combining with brain connectivity and neuron spontaneous activity with small ball athletes in different training stages.The main tasks are as follows:1.Calculated the dynamic functional connectivity density for small ball elite athletes,student athletes and healthy controls based on functional magnetic resonance imaging.The results were analyzed by variance analysis,k-means cluster,Markov Chain and so on.It was found that the three groups had significant differences in bilateral inferior frontal gyrus orbital part,right lingual gyrus,bilateral postcentral gyrus and left paracentral lobule.At the same time,the temporal variability of functional connectivity density of professional athletes in most differential regions has a tendency to return to healthy controls.It indicated that small ball game training would cause plasticity changes to the brain,and the dynamic brain adaptation fluctuations caused by different training stages were different.2.Calculated the dynamic amplitude of low-frequency fluctuations for table tennis professional athletes,college student athletes and normal controls based on functional magnetic resonance imaging.The results of variance analysis showed that there were significant differences in the left cerebellum 8 and left temporal gyrus,and showed that the variation of low-frequency fluctuations of college student athletes were greater than professional athletes,and professional athletes were greater than normal controls,and professional athletes had a tendency to return to normal control group.The correlation between the dynamic amplitude of low-frequency fluctuations in the different brain regions and the training time was found to be negatively correlated at the initial stage of training,but not at the later stage of training.Showed that table tennis can bring changes in brain plasticity,and in the early stage of training will increase the movement and vision related brain area of spontaneous activity variability,to the later stage of training spontaneous activity tends to stabilize,showing the characteristics of brain resource saving.3.Using Fourier transform for time-frequency conversion of functional magnetic resonance imaging data.Then calculated the information entropy of different window width and different step time window to evaluate the selection of two parameters(window width and step size)of sliding window correlation method.The results showed that the false fluctuation can be basically eliminated after the window width is more than 50 TR,and there was a more balanced state between the dynamic and containing semaphore of the time window after 68 TR,and the time window of about 90 TR width was basically lost dynamics.The law of information entropy increasing with step size is not obvious,but the general trend is increasing.But for our problem,it was better to select a small numerical step size where appears peak value.