Advanced Machine Learning for Preschooler Magnetic Resonance Imaging Analysis in Classification of Anxiety Disorders

Numerous individuals suffer from anxiety disorder. Treatments for anxiety usually involve psychologists and counselors based on qualitative data from interviews and conversations to make an educated guess to classify their anxiety. We built a quantitative method for the diagnosis of anxiety, which can be used by psychologists and doctors to obtain accurate data to treat it. The data were obtained from OpenNeuro and Magnetic Resonance Imaging (MRI) scan images of preschoolers with different types of anxiety: generalized anxiety and separation anxiety. These data were used to train machine learning models: Support Vector Machines (SVMs), decision trees, and Logistic Regression. The dataset consisted of MRI images. The method was refined until the desired accuracy was obtained. The model can be used to diagnose anxiety disorders for patients to be treated with a personalized approach.