Exploring Machine Learning Models for Schizophrenia Detection: A Systematic Review

Abstract

Schizophrenia is a neurological disorder that is associated with several genetic, environmental, and neurobiological factors. The various prediction, detection, and classification techniques that can be used in order to catch the disease in early stages are noted in this paper. Some of the common machine learning algorithms, such as convolutional neural networks (CNNs), support vector machines (SVMs), random forest, logistic regression, and also certain models that have been developed, particularly those by the PRONIA project and ENIGMA consortium, are compared in terms of their results and accuracy in schizophrenia detection. The aim is to identify which of these machine learning models performs the best. While a lot of developments have been made, there is much more that can be done in this particular field of psychosis. Additionally, a potential new model is proposed for future work.