ABSTRACT
Objective:To know the clinical characteristics, seasonal pattern and influencing factors of atypical depression(AD) patients.Methods:A total of 203 depressed outpatients of Peking University Sixth Hospital from January 2021 to August 2021 were included.They were assessed with demographic questionnaire, inventory of depressive symptomatology self-report(IDS-SR30) and seasonal pattern assessment questionnaire(SPAQ). According the score of IDS-SR30, all patients were classified as atypical depression(AD) and non-atypical depression(non-AD). The data were analyzed by t-test, non-parametric test and Logistic regression using SPSS 26.0 software. Results:The prevalence of AD among depressed patients was 36.0% (95% CI=29.3%-42.6%). The IDS-30 score of the AD group was (41.59±10.59), and IDS-30 score of the non-AD group was (36.08±13.17), and the difference between the two groups was statistically significant ( t=3.062, P<0.05). The global seasonal score of the AD group was 6 (3, 9), and 17.8% of the AD group had seasonal pattren.The global seasonal score of the non-AD group was 5 (3, 8), and 14.6% of the non-AD group had seasonal pattern.There was no significant difference in the global seasonal score and the proportion of seasonal pattern between the two groups ( Z=0.389, χ2=0.359, P>0.05). Depression patients who were females ( β=1.08, OR=2.95, 95% CI=1.32-6.59, P<0.05), low self-evaluation ( β=0.82, OR=2.27, 95% CI=1.12-4.59, P<0.05)and psychomotor retardation ( β=0.93, OR=2.54, 95% CI=1.33-4.85, P<0.05) were more likely to be diagnosed as AD, and depression patients having mood variation ( β=-0.94, OR=0.39, 95% CI=0.19-0.81, P<0.05) were more likely to be diagnosed as non-AD. Conclusion:Women, low self-evaluation, psychomotor retardation and unobvious mood variation can predict and help to diagnose atypical depression in depressed patients.
ABSTRACT
Objective To explore the program of convolutional neural networks for the diagnosis of schizophrenia and evaluate its effects. Methods Using the convolutional neural network,the training model was trained in the lead data of 138 normal people and 183 schizophrenic patients,and the model was valida-ted by 20-fold cross-validation. Results The true positive rate of schizophrenia prediction using the convolu-tional neural network training model was 0. 749, the false positive rate was 0. 275, and the accuracy was 0. 738. Conclusion This model can achieve a strong diagnostic ability for patients with schizophrenia. Therefore,convolutional neural network for the diagnosis of schizophrenia will become an important research direction in the future.
ABSTRACT
Objective To explore the sensitivity and specificity of near-infrared spectroscopy ( NIRS) in the diagnosis of depression. Methods From March 2013 to August 2013,62 patients with de-pression and 70 normal controls were collected from Peking University Sixth Hospital. Optical Tomography System (52 channels) was used to collect the NIRS data during the Verbal Fluency Test (VFT),and the number of words produced during VFT task was recorded. The wave analysis was performed by a professional psychiatrist. Results There was statistical difference in the number of words produced during VFT task be-tween the patients with depression (8.65±0.49)and control group(10.19±0.43) ( t=2.385, P<0.05). Through the wave analysis of NIRS to test patients with depression,the results demonstrated that the sensitivi-ty was 66.1% and the specificity was 91.4%. Conclusion The results of NIRS test display high specificity in the diagnosis of depression,which can be used as an objective index for clinical auxiliary diagnosis.
ABSTRACT
Objective To explore the effects of emotion on cognitive function in patients with depression.Methods Self-rating anxiety scale(SAS),self-rating depression scale (SDS) and the typical one-back task were done by 397 patients with depression in Peking University Sixth hospital out-patient and hospitalization during 2012 June to 2013 October.Results Both of the SAS standard score (51.81 ± 11.50) and SDS standard score (62.94 ± 13.06) were significantly related (r=0.125,P＜0.05.r=0.176,P＜0.01) to the reaction time of one-back task ((590.27±213.96)ms),and the correlation between SAS standard score and SDS standard score was significant (r=0.682,P＜0.01).The results of regression analysis suggested that,only SDS score could predict the reaction time of one-back task.Conclusion The emotion of anxiety and depression in patients with depression are correlated with cognitive function,and the emotion of depression is the main factor to affect the cognitive function.