Android Patient Robot Simulating Depressed Patients for Diagnosis Training of Psychiatric Trainees

This paper proposes a depression patient robot using the android robot. The purpose of this patient robot is to offer standardized and reproducible diagnosis training and education to psychiatrist trainees by using robot technologies instead of human standardized patients. Furthermore, it can be expected that psychiatrist trainees are able to experience realistic diagnosis as if they have an interview with a real human standardized patient by taking advantage of android robot technologies. In the proposed system, as the first step, the patient robot mimics unipolar depression, because it is a typical psychiatric disease. The robot expresses symptoms of unipolar depression by exploiting its communicative functions such as facial expressions, eye and head movements, and speech. Robot\´s speech and communicative behavior are controlled remotely by an operator according to a diagnosis scenario. The diagnosis scenario, question and answer process between an interviewer and the patient robot, is prepared based on the "Structured Interview Guide for the Hamilton Depression Rating Scale" (SIGH-D) which are widely used for diagnosis of depressed symptoms. Finally, some psychiatrists evaluate the proposed system through a diagnosis experiment.