A research designed and carried out by Üsküdar University Faculty of Medicine, Psychiatry and Hospital Department, Psychiatry Program shared effective results of machine learning methods in the diagnosis and classification of psychiatric disorders. The study showed the potential of EEG data as discriminative biomarker for the diagnosis of different psychiatrist classification.
In the research carried out by Üsküdar University Faculty of Medicine Psychiatry and Hospital Department, Psychiatry Program, the dataset of EEG measurements of individuals diagnosed with bipolar disorder, attention deficit and hyperactivity disorder (ADHD), depression, obsessive compulsive disorder (OCD), opioid addiction, post-traumatic stress disorder (PTSD), schizophrenia and healthy individuals was used as biomarker, and an anticipation model was developed to support the diagnosis with computer-aided machine learning methods.
Can psychiatric diseases be diagnosed, classified with EEG data?
The findings obtained within the scope of the study showed that it can be highly predicted whether a person has a psychiatric disease or not by using EEG data as a biomarker. According to this, it is possible to analyze whether the person who consults a physician with a complaint is ranked among the psychiatric disease class with EEG measurement once evaluated with broad strokes and it can be distinguished by its discriminative features among other disease categories.
Prof. Nevzat Tarhan: “Machine Learning will contribute to the field of psychiatry”
President of Üsküdar University Prof. Nevzat Tarhan who summarized the research said that “W When trying to differentiate between numerous and diverse disease categories, it may be claimed that some diseases (ADHD, depression, schizophrenia) can be distinguished better. Considering the findings, it is anticipated that the analyzes obtained as a result of this study will contribute to the studies to be conducted using machine learning in the field of psychiatry. Especially anticipation models developed with new generation in-silico methods reveal significant clinical results. Although old generation surface learning algorithms and statistical models contributed to the literatures, especially deep learning based new generation learning algorithms that can distinguish features contribute to the development of studies in the neurosciences, diagnostic and prognostic process, early diagnosis and correct treatment process and physicians. “.
Prof. Nevzat Tarhan stated that: “Physicians follow-up a symptom-based approach in the diagnosis of psychiatric diseases. According to this approach, a process based on internationally valid diagnostic tools such as The Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases (ICD), patient reports and the observation and experience of the physician is monitored. As in other fields of medicine, the search for biomarkers that can be used in processes related to diseases continues in psychiatry and various researches are carried out in this field.
The study that Faculty Member of Üsküdar University Faculty of Medicine, Psychiatry Department Prof. Cumhur Taş also participated, was published on ‘International Journal of Medical Informatics’. The dataset containing electroencephalogram (EEG) measurements of patients diagnosed with psychiatry diseases (550 patients) obtained within the scope of the study were analyzed by machine learning methods, and diseases were classified according to the models obtained.
