Schizophrenia classifying method based on task switch test behavior data

Abstract

The invention provides a schizophrenia classifying method based on task switch test behavior data, in order to solve the problems that the present schizophrenic characteristic data acquisition is highin cost, the analysis is complex, the quick, objective and accurate response cannot be made, and the like. The method comprises the following steps: establishing a task switch test experimental procedure; acquiring a task switch test experimental data of a tested user, namely, response time and accuracy; calculating the response time difference related to the execution control according to average response time; utilizing the task switch test experimental data and the response time difference related to the execution control as features for training the optimal self-adaption promotion classifier; using a cross validation method for verifying the optimal self-adaption promotion classifier. According to the schizophrenia classifying method, 90% of classification accuracy, 100% of sensitivity and 81% of specificity can be achieved; the area under the working characteristic curve of the tested user is 0.89; the method is capable of providing objective and effective assistance in identifying the schizophrenic patient and has a huge potential application value in the research field of mental diseases.

Description

technical field [0001] The invention relates to a method for classifying schizophrenia based on behavioral data of a task switching test. Background technique [0002] Schizophrenia is a clinical syndrome composed of a group of symptoms, and it is a multifactorial disease. Although the current understanding of its etiology is not very clear. Therefore, the early detection and early diagnosis of schizophrenia is an urgent problem that the whole society needs to solve. The current diagnostic methods for schizophrenia usually use assessment scales and image analysis. The most commonly used scales are Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS). For scale analysis, It's too subjective. Neuroimaging techniques mainly include computed tomography (CT) and magnetic resonance imaging (MRI). At present, many neuroimaging studies have found that the relevant brain regions of schizophrenia patients have abnormal gray matter and white matter ...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the existing schizophrenia clinical scale has strong subjectivity and large errors, and the neuroimaging-based classification method has high image data acquisition cost and complicated analysis, it is difficult to check and diagnose schizophrenia. It is impossible to give quick, objective and accurate answers to the judgment of schizophrenia, and a classification method of schizophrenia based on behavioral data of task switching test is provided

Images