Correlation Analysis Of College Students’ Depressive Tendency And Mental State

Depression is the most vital public mental health problem around the world,and it does great harm to public health.Therefore,accurate screening and diagnosis,effective intervention and treatment of depression are of great significance.However,due to the complex causes of depression,the current recognition mainly depends on the scale and the doctor’s diagnosis,which has the problems of difficult diagnosis,lack of objectivity and low accuracy.Thus,how to accurately identify depression through objective physiological indicators is particularly important.Therefore,the purpose of this study is to investigate the impact of depression tendency on the mental state of college students based on fNIRS,EEG and ECG techniques,and to explore whether there are physiological indicators that can effectively distinguish depression states,so as to identify and diagnose depression more objectively and accurately,and help improve the mental health of college students.In this study,74 college students were used as subjects.First,resting state experiments were carried out,and DLPFC brain activation signals,EEG,and ECG signals were collected simultaneously,and then a paper-and-pencil test was performed.Using The Center for Epidemiologic Studies Depression Scale（CES-D）to measure the degree of depression of the subjects,and Raven’s Advanced Progressive Matrices（RAPM）for measuring the level of fluid intelligence of the subjects.According to the data screening criteria of each modality,there were 65 valid subjects for fNIRS,67valid subjects for ECG,and 52 valid subjects for EEG.Based on the depression cut-off points recommended by CES-D,subjects with CES-D score≥16 points were divided into high depression tendency group,and the subjects with CES-D score<16 points were divided into low depression tendency group.First,descriptive statistics were performed on the questionnaire scores of the subjects to understand the current depression tendency and fluid intelligence level of the subjects.Then the experimental data was processed for each mode:for the fNIRS data,the Homer3 toolkit in MATLAB 2017b was used to convert the brain activation signal into the Hb T value according to the modified Beer-Lambert law for pairwise Pearson correlation calculation to form the functional connectivity matrix,then perform Fisher’s Z transformation and the converted Hb T functional matrix value was used to represent the level of DLPFC brain functional connectivity.For the EEG signal,first visually inspect it through EEGLAB and remove the noise,and then use the fast Fourier transform to decompose the EEG signal into different frequency bands through MATLAB 2017b:δ（1-4Hz）,θ（4-8Hz）,α（8-13Hz）,β（13-30Hz）and calculate the absolute power of each frequency band and the total absolute power of 1-30Hz,then divide the absolute power of different frequency bands by the absolute power of 1-30Hz to calculate the relative power ratio,and finally according to the formula“10*log₁₀（relative power ratio of each frequency band）”calculates the relative power value of each frequency band after logarithmic transformation.For the ECG signal,the heart rate and heart rate variability indexes are calculated first,and then the heart rate variability is analyzed in time domain.The indexes used include SDNN,RMSSD,and p NN50.In order to explore the influence of depression tendency on the mental state of the subjects,the obtained Hb T function matrix value,the relative power value of each frequency band of EEG,heart rate and heart rate variability indexes were descriptive statistics based on SPSS 26.0 software,and were biased with the questionnaire scores of the subjects.related analysis.After the above analysis,the main results are as follows:（1）There is little difference in fluid intelligence scores among the subjects.（2）There was an opposite correlation trend between the mean value of DLPFC brain functional connectivity and RAPM scores in the high and low depression tendency groups(r _{high depression subjects}=0.158,r _{low depression subjects}=-0.245).（3）The EEG indexδ/1-30Hz of the subjects with high and low depression tendencies had an opposite trend to the CES-D score(r _{high depression subjects}=0.085,r _{low depression subjects}=-0.501*).（4）There was a significant positive correlation between SDNN and CES-D scores of all subjects（r=0.321**）.Based on the above results,the following conclusions can be drawn:（1）The fluid intelligence levels of the subjects with high and low depression tendencies are similar,but the DLPFC brain functional connectivity of the subjects with high depression tendencies is weaker,indicating that depression tendencies have no effect on the fluid intelligence levels of the subjects and there is an impact on the DLPFC brain functional connectivity level of the subjects.（2）There was a significant negative correlation between the relative power of theδfrequency band and the score of the depression scale in subjects with low depression tendencies,indicating that the stronger the activity of theδfrequency band,the lower the tendency to depression;at the same time,the relative power of theδfrequency band of all subjects was significantly positively correlated with the level of fluid intelligence.Correlation indicates that the stronger the activity of theδfrequency band,the higher the fluid intelligence level of the individual.（3）The depression scale scores of all subjects were significantly positively correlated with the heart rate variability index SDNN,indicating that the greater the degree of heart rate variability of the subjects,the degree of depression may also increase accordingly,and it is suggested that SDNN can be used as an objective physiological index to measure Individual’s propensity to depression.（4）When identifying an individual’s depression tendency,it can be judged by integrating fNIRS,EEG,and ECG technologies comprehensively from different angles,like using the physiological index SDNN to make a preliminary judgment,and further distinguishing in combination with EEG signals and brain functional connectivity levels to improve the identification of depression tendency efficiency and accuracy.