Preliminary Study on Model Construction for Estimating Mental Illness and Developmental Disorders

In recent years, the number of patients with mental disorders and developmental disabilities has been increasing. Current diagnostic methods for these patients are mainly interviews between clinicians and patients, which is regarded as subjective evaluation that may cause instability. Therefore, it is necessary to support the clinician's diagnosis by objective evaluation. We conducted preliminary study on constructing a model to support the diagnosis from electroencephalogram (EEG) applying deep learning. To construct the model, EEG data was collected from six people with mental illness or developmental disabilities and three people without these disabilities during resting with eyes closed. The collected EEG data were used as explanatory variables. A binary value, which means the group with or without mental disorders and developmental disabilities, were used as objective variable. We trained the model and verified the model with cross validation which training and test data do not include the same participant's data. The results of the accuracy verification suggested that it is possible to construct a model with an average accuracy of about 70%.