The Mechanism Of Cognitive Impairment And The Development Of Multi-dimensional Identification Model Of College Students At Ultra-high Risk For Psychosis

Objectives: To investigate the epidemiological characteristics of Ultra-High Risk(UHR)syndrome for psychosis among Chinese college students,to explore the features of cognitive impairment and its neuroimaging mechanism in UHR subjects,and to establish a multidimensional UHR recognition model based on cognitive function and neuroimaging indicators.Methods: A total of 23063 college students from 15 universities in Shanghai,Nanjing and Nanchang were investigated in this study.A 3-stage method that consists of scale assessment,telephone interview and the Structured Interview for Prodromal Syndromes(SIPS)was conducted to detect UHR subjects and to investigate the epidemiological characteristics of UHR syndrome.The detection rate of UHR and the distribution characteristics of prodromal symptoms in college students in China were explored,and the application value of relevant screening tools in this group was verified.In addition,the MATRICS consensus cognitive battery(MCCB)data,structural and resting state-functional magnetic resonance imaging data of 35 UHR subjects and 31 agematched and gender-matched healthy controls were collected to analyze the characteristics of cognitive impairment in UHR subjects and its relationships with the gray matter volume and the functional connectivity of brain regions concerning the psychopathology related brain circuits.Also,we analyzed the mediating effect of cognitive function on brain imaging alterations and prodromal symptoms.Finally,we further used support vector machine(SVM)algorithm to establish four UHR recognition models based on different data modality,with demographic,cognitive and brain image markers of UHR subjects(n=35)and healthy controls(n=31)defined as classification features.The sensitivity and specificity of the models were measured.Results: This study found that:(1)The detection rate of UHR in this sample was0.3%,the peak age range of UHR subjects was 17-20 years old,and the age range was slightly wider in women than in men.In the detection of UHR subjects,the prediction accuracy of the distress score of the Prodromal Questionnaire-Brief Version(PQ-B)was the highest.With a cut-off point of 46,a sensitivity of 90.3% and a specificity of 81.2%could be obtained.In the analysis of the significance of the score of each scale item in screening UHR subjects,the key items of PQ-B were items No.1,No.4,No.12,No.13,No.15,No.17,No.18,No.21,those of the questionnaire of Personality QuestionnaireSchizotypal Personality Disorder were items No.1,No.3,No.4,No.5,No.6,No.8,and those of the Negative,Disorganization,and General Symptoms Scale were items No.2,No.13,No.14,No.15,No.16,and No.19.(2)Speed of processing and verbal learning were the main two dimensions of cognitive impairment in college students with UHR.In UHR subjects,the gray matter volume of putamen increased significantly and the functional connectivities between multiple brain areas in cerebellum,thalamus,and cortex decreased,so as the functional connectivities between hippocampus and parietal lobe,temporal lobe,precuneus,and so as the functional connectivities between putamen and parietal lobe,temporal lobe,precuneus,insula,and cerebellum.In the UHR group,the working memory was positively correlated with the gray matter density of the left Rolandic operculum,the left insula,the left superior temporal gyrus and the left inferior frontal gyrus,triangular part.The functional connectivity strength of the left fusiform gyrus and the right precentral gyrus was correlated with the speed of information processing and attention/vigilance of the UHR subjects.Moreover,the speed of information processing had a significant complete mediating effect between the functional connectivity and the motor disorder symptoms of the UHR subjects.(3)The accuracy,sensitivity and specificity of the SVM classification model based on f MRI functional connectivity data in the recognition of UHR syndrome were 83.1%,79.4% and 89.1%,respectively,which were significantly higher than the multidimensional model based on gray matter volume,demographic and cognitive data.Moreover,across multiple data types,functional connectivity is far more important to the classification of UHR and HC than other characteristics,including demographic information and cognitive function.Conclusions: The present study indicates that:(1)The detection rate of UHR syndrome in Chinese college students is about 0.3%,and it is mainly prevalent in the age group of 17-20 years old.The 3-stage method can be applied to the epidemiological investigation of UHR in a large sample of general population.Delusional ideas,perceptual abnormalities,suspiciousness,schizotypal personality traits,social anhedonia,dysthymia,and impaired tolerance to normal stress are significant prodromal symptoms that contribute to the recognition of UHR.(2)The cognitive impairment in UHR subjects mainly manifest in the decrease of information processing speed and the decline of verbal learning ability.Among them,the speed of information processing was significantly correlated to the aberrant functional connectivity between the left fusiform gyrus and the right precentral gyrus,and this altered functional connectivity may indirectly lead to the development of motor disorder symptoms in UHR subjects by affecting the speed of information processing.(3)The recognition model based on brain functional connectivities can effectively distinguish UHR subjects from healthy controls with high specificity.Compared with data dimensions such as gray matter volume,demographic data and cognitive function,functional connectivity features are of high value for UHR recognition.Adding data dimensions on this basis cannot improve the accuracy of the model.