Analysis Of Multivariate PSG Data In REM Sleep Behavior Disorder And Parkinson’s Disease

Part one:Objective:Parkinson’s disease(PD)is the second most common neurodegenerative disease.In recent years,with the trend of population aging,the incidence of PD disease is increasing year by year.Rapid eye movement(REM)sleep behavior disorder(RBD)carry a high specific risk for developing a neurodegenerative disorder of alpha-synucleinopathy(e.g.Parkinson’s disease,dementia with Lewy bodies and multiple system atrophy)within years to decades.Early diagnosis of PD remains a challenge to date.As one of the most common non-motor symptoms in PD,RBD is of great importance in the early diagnosis of PD.We collected overnight sleep data via polysomnography(PSG),aimed to investigate the sleep structure and the presence of potential early manifestations of PD,in RBD patients,PD patients and normal controls,tried to provide an early diagnosis and identify therapeutic targets.Method:The study was successively included the case group in the Department of Neurology of Shaanxi Provincial People’s Hospital from January 2018 to October 2020;at the same time,other hospitalized in the Department of Neurology of Shaanxi Provincial People’s Hospital in the same period were included as the control group,both groups of patients met the inclusion and exclusion criteria.In this study,the clinical data of the three groups,including age,gender,were analyzed and compared by case-control.Patients were divided into control,RBD and PD groups according to the results of polysomnography(PSG),based on the diagnostic criteria of the International Classification of Sleep Disorders(ISCD)-II criteria.Polysomnography were taken to collect the sleep parameters such as sleep efficiency(SE),wake time after sleep onset(WASO),sleep latency(SL),apnea hypopnea index(AHI),periodic limb movement index(PLMI),arousal index(AI)obstructive sleep apnea(OSA),periodic limb movements in sleep(PLMS),rapid eye movement sleep latency,each sleep period and the percentage of each sleep period during sleep,and which were analyzed and compared.The difference was statistically significant(P<0.05).All data were analyzed by SPSS 18.0 statistical software.Result:1.According to the inclusion and exclusion criteria,44 subjects were included,including 13 cases in the control group and 31 patients in the case group.There was no significant difference in age and gender among the three groups.2.The SE,the percentage of REM(R%),the percentage of N2(N2%)in PD group was significantly lower than that in control group(P < 0.05).There was no significant difference between the three groups in WASO and N2%.3.There was no significant difference between the three groups in OSA,PLMS events and number of awakenings.4.The SL in PD group was significantly higher than that in control group(P < 0.05).5.The REM-SL in PD group was significantly higher than that in control group(P < 0.05).6.The percentage of N1(N1%)in PD group was significantly higher than that in control group(P < 0.05).7.There was no significant difference between the three groups in AI,AHI and PLMI.Conclusion:1.Patients with PD have significant and objective sleep disorders with clinical heterogeneity and various manifestations.2.RBD is a prodromal state of PD,with RBD a kind of intermediate state between controls and PD.3.The application of PSG will greatly help detect early clinical and even preclinical cases and provide a critical opportunity for neuro-protective therapies and optimize diagnosis and treatment mode.Part two:Objective:Parkinson’s disease(PD)is the second most common neurodegenerative disease.In recent years,with the trend of population aging,the incidence of PD disease is increasing year by year.Rapid eye movement(REM)sleep behavior disorder(RBD)carry a high specific risk for developing a neurodegenerative disorder of alpha-synucleinopathy(e.g.Parkinson’s disease,dementia with Lewy bodies and multiple system atrophy)within years to decades.Early diagnosis of PD remains a challenge to date.As one of the most common non-motor symptoms in PD,RBD is of great importance in early diagnosis of PD.We collected overnight electroencephalogram via polysomnography(PSG),aimed to find a suitable method of EEG feature extraction and apply it to disease diagnosis modeling,tried to provide an early diagnosis and identify therapeutic targets.Method:The study was successively included the case group in the Department of Neurology of Shaanxi Provincial People’s Hospital from January 2018 to October 2020;at the same time,other hospitalized in the Department of Neurology of Shaanxi Provincial People’s Hospital in the same period were included as the control group,both groups of patients met the inclusion and exclusion criteria.In this study,the clinical data of the three groups,including age,gender,were analyzed and compared by case-control.Patients were divided into control,RBD and PD groups according to the results of polysomnography(PSG),based on the diagnostic criteria of the International Classification of Sleep Disorders(ISCD)-II criteria.EEG features of four groups were extracted and observed.Result:1.According to the inclusion and exclusion criteria,44 subjects were included,including 13 cases in the control group and 31 patients in the case group.There was no significant difference in age and gender among the three groups.2.The approximate entropy method was used to extract the features of EEG signals.The NI,N2,N3,R,W among three groups was significantly difference(P < 0.05),with all pairwise comparisons significantly different.3.The fuzzy entropy method was used to extract the features of EEG signals.The NI,N2,N3,R,W among three groups was significantly difference(P < 0.05),with all pairwise comparisons significantly different,except the R among the PD and RBD groups4.The wavelet packet decomposition was used to extract the energy features of EEG signals.At(3,0),the N1,N2 and R wavelet packet energy value(NWPEV)in PD group was significantly lower than others groups.There was no significant difference between the three groups in N3 and W NWPEV.And at(3,4),the NWPEV of PD group increased significantly in each sleep stage,while the other two groups had no significant change.Conclusion:1.This study can be found that approximate entropy and fuzzy entropy has better performance in characterization PD EEG signal.2.The increased energy values are likely to be a biomarker of Parkinson’s disease in wavelet packet decomposition(3,4).3.Several potential features,including approximate entropy,fuzzy entropy and wavelet decomposition that can measure and quantify the complexity or irregularity of EEG signals,and which can be used in the modeling of EEG diagnosis of diseases in the future.