Efficacy of the Mobile Application for Prediction and Prevention of Mood Episode Recurrence Based on Machine Learning

Efficacy of the Mobile Phone Application (Circadian Rhythms for Mood) for Prediction and Prevention of Mood Episode Recurrence in Mood Disorders Based on Machine Learning of Daily Digital Phenotype Variables : A Sham-controlled Randomized Clinical Trial

This study was designed to evaluate the efficacy of the mobile application named Circadian Rhythm for Mood (CRM), which was developed to prevent recurring episodes of mood disorders (major depressive disorders, bipolar disorders type 1 and 2) based on machine learning.

Inclusion Criteria:

• Male and female patients, 19-70 years old
• Diagnosis with Bipolar I disorder, Bipolar II disorder, Major depressive disorder based on DSM-5 criteria, in euthymic state for more than two weeks at the time of the recruitment
• Android smartphone users, capable of installing and executing the CRM application
• Consent to wear wearable device (Fitbit) continuously and synchronize and backup data regularly

Exclusion Criteria:

• Patients who have not experienced major depressive, manic, or hypomanic episode in the last two years
• Patients who are difficult to specify mood episode or evaluate symptoms of mood episode independently due to personality traits (borderline personality trait, cyclothymic temperament, etc.)
• Patients with degenerative neurological disorders (Parkinson's disease, dementia, Huntington's disease, etc.), neurodevelopmental disorders (intellectual disorder, autism spectrum disorder, down syndrome, etc.), epilepsy, severe traumatic brain damage, stroke, and other brain neurological disorders
• Inmates or patients who are forced into custody for the treatment of mental or physical illness (non-voluntary isolation or hospitalization)
• Patients with difficulties in understanding the objectives and process of the study and the potential benefits and risks of participating in the study