Detection of First-Episode of Schizophrenia Brain MRI Images Using Random Forest Classifier

Abstract

Schizophrenia (SCZ) is a mental disorder, that is, persons loss their normal life and they supposed to live in an imaginary world of hallucination or delusions. Segmentation of brain tissue is an important task for the accurate detection of the disease SCZ. Here, proposing a novel method to segment the brain tissue and obtained the dominated features for the detection. Medical imaging is used and is a technique that processes and creates visual representation of the interior of a body for clinical analysis and medical intervention. Here, integrating two segmentation procedures is based on probability distribution and various intensity level heuristic segmentations, which make use of swarm optimization. A better result can be made using the extracted features of gray matter. ANOVA was used to analyze the statistical performance. The result shows that it yields accuracy 99.09%, sensitivity 100% and specificity 98.14% for schizophrenic detection with the use of random forest classifier.