Multimodal Validation of Facial Expression Detection Software, for Real-time Monitoring of Affect in Patients with Suicidal Intent
Refereed Conference Meeting Proceeding
High risk for suicide is typically assessed by clinicians using questionnaires and interviews. Although useful in a wide range of clinical settings, this assessment approach has many disadvantages e.g., misinterpretation of subtle differences in meanings of words used in emotional scales, objective and subjective biases and difficulties in reliably assessing the intensity of the emotion. Most significantly, these cues can be missed with catastrophic consequences. In this context, we suggest that novel, non-intrusive facial affect detection technology could play a role in the clinical evaluation of suicidal ideation. We report on the acquisition of discrete emotional states (i.e., fear, sadness, joy, anger, disgust and surprise) while the patient is participating in a standardised task utilising the presentation of emotionally challenging images.
24th European congress of Psychiatry EPA2016
Digital Object Identifer (DOI):
Dublin City University (DCU)
Open access repository: