Analysis And Research On Chronic Insomnia Among College Students In Beijing Based On HRV And Establishment Of Deficiency-excess Syndrome Differentiation Mode

Insomnia is a very common sleep disorder.It refers to the subjective experience of inadequate sleep duration and/or quality despite adequate sleep opportunities and environments,and can affect various social functions during the day.The main manifestations are difficulty falling asleep,multiple dreams,wakefulness,decreased sleep quality and decreased total sleep time,accompanied by varying degrees of daytime dysfunction,such as fatigue,irritability,social difficulties,memory decline,reduced work efficiency and so on.It has a great impact on people’s quality of life,mode of production and social and public security,and has brought heavy medical consumption to the country and society.So insomnia is both a medical problem and a social problem.In recent years,it has become a hot issue of common concern to researchers at home and abroad.At present,the treatment of insomnia mainly has drug therapy and non-drug therapy.Western medicine drug therapy has the characteristics of convenient use,accurate curative effect and wide application,but long-term use may produce drug resistance,dependence and addiction.Once stopped,symptoms will rebound and withdrawal effect,resulting in repeated attacks and prolonged course of disease.TCM has a long history of understanding and treating insomnia.Based on the overall concept,TCM treats insomnia based on syndrome differentiation,has various treatment methods,no addiction and hangovers like effects,high treatment efficiency,and little toxic and side effects,which has many congenital advantages.The mastery of the traditional four diagnostic methods is limited by the doctors’long-term repeated practice and condensed sublimation,and due to the influence factors such as different individual experience,different cognitive level and different learning background,the clinical syndrome differentiation among different doctors is not unified,which restricts the popularization and promotion of TCM characteristic therapy to a certain extent.Therefore,it is particularly important to carry forward the essence of traditional medicine,absorb the modern monitoring technology,obtain the objective physical signs information of insomnia patients from multiple perspectives,and establish the objective,standardized,quantifiable standard of TCM syndrome differentiation and classification of insomnia that can be carried out in a wide range of clinical and scientific research.Therefore,the development of intelligent research on TCM diagnosis is of great practical significance for promoting the modernization of TCM,enriching TCM diagnosis and treatment methods and technologies,promoting the combination of medical and industry,and innovating the community modernization of medical services.With the above problems in mind,this study attempted the initial design and exploration of a TCM insomnia syndrome differentiation and classification model based on HRV as the core physiological parameters,including the following parts:Objective1.Explore the distribution of TCM syndromes of chronic insomnia among college students;2.Different physiological parameters related to sleep were compared between college students with chronic insomnia and healthy people,and the differences of sleep physiology,sleep structure and sleep behavior between the two groups were compared.3.SVM,KNN,DT,RF,ET,LightGBM and other algorithms were used to construct the syndrome differentiation and classification model of insomnia TCM deficiency syndrome and hot syndrome based on HRV as the core parameters,and the optimal term was screened out.MethodsA cross-sectional study was conducted on 223 college students studying in BUCM from September 2020 to April 2022,including 183 insomnia subjects and 40 healthy subjects.This study was approved by the Medical Ethics Committee of BUCM and followed up the whole process.The subjects voluntarily participated and signed the informed consent and met the inclusion criteria.The TCM syndrome questionnaire of insomnia,Pittsburgh Sleep Quality Index(PSQI)and Self-rating Sleep Status Scale(SRSS)were completed.The HRV-based non-contact heart rate and respiration recorder was used to monitor sleep quality and collect physiological data,and the physiological data were recorded.Through the combination of"looking,smelling,asking and cutting",the insomnia was divided into false and real according to the existing TCM classification standard.Among them,liver depression and fire,phlegm and heat internal disturbance are empirical evidence,Yin deficiency and fire,heart and spleen deficiency,and gallbladder and qi deficiency are deficiency syndrome,and then establish and perfect database.The differences of sleep related physiological parameters between insomnia subjects and healthy subjects were analyzed.Compare the distribution law of TCM syndrome of insomnia population;Based on the classification of TCM deficiency and reality,the machine learning algorithm was introduced to construct the TCM deficiency and reality classification model of insomnia with HRV as the core physiological parameter.Results1.By comparing the physiological parameters related to sleep in healthy people,the following changes exist in college students with chronic insomnia:(1)Sleep physiology:low LF/HF,high average heart rate and RR_max,low RR_min,low immunity index and high risk of infection;(2)Sleep structure:low sleep duration,high time required to fall asleep,low light sleep duration,low REM sleep duration,low deep sleep duration and proportion,high awake duration and proportion;(3)Sleep behavior:the number of times of leaving the pillow and the height of the hour,the number of body movements was high,and the average body movement time was low.2.Distribution of insomnia syndrome:deficiency syndrome(116 patients,63.4%)and positive syndrome(67 patients,36.6%).The order from high to low was as follows:Yin deficiency with fire flourishing(52 people,28.4%)>heart and spleen deficiency(44 people,24.0%)>phlegm-heat internal disturbance(37 people,20.2%)>Liver stagnation with fire flourishing(30 people,16.4%)>heart and bile qi deficiency(20 people,10.9%).3.Compared the effectiveness of the six machine learning models for insomnia syndrome differentiation,S VM was the best,with Accuracy of 0.944444 and AUC of 0.946488,showing good diagnostic value.Conclusion1.There are significant differences in sleep physiology,sleep structure and sleep behavior between college students with chronic insomnia and healthy people.2.The chronic insomnia of college students is mainly due to deficiency syndrome,among which Yin deficiency and fire burning syndrome are more common.3.It is feasible to establish a syndrome differentiation and classification model of insomnia based on HRV as the core physiological parameters by machine learning algorithm.