Machine Learning-Based Detection of Children at Risks for Mental Disorders

Schizophrenia imaging is a noninvasive way of monitoring the electrical activity in human brain. The Schizophrenia Database plays a major role in identifying young people who may have mental health problems. However, the analysis of schizophrenia data is challenging because of its high dimensionality and complexity. This research study describes an innovative approach by combining PCA with the Bayesian algorithm's probabilities for detecting mental illnesses such as schizophrenia, anxiety, and stress disorders. Here, PCA is initially used to reduce the residual features of schizophrenia. Following that, the value-weighted negative Bayes algorithm is used to sort the schizophrenia data based on the probability of schizophrenic patients. The proposed model was evaluated using a schizophrenia database that included information on both children with and without mental health problems. The results of this research study show that the proposed approach is effective in detecting mental health problems.