Learning based visual engagement and self-efficacy

Self-help web interventions for mental health effectively follow a one-size-fits-all approach lacking the personalization of regular psychotherapy sessions and the effectiveness associated with the treatment. A scalable adaptive person-centered approach is therefore essential to non-invasively monitor symptom severity, enhance coping capabilities and increase engagement levels for maximal impact. In this work, we propose a novel approach to empower mental trauma patients, improve outcomes and EASE their suffering while reducing health-care costs. We develop machine learning algorithms that use both voluntary and involuntary feedback encapsulating the interactions of brain and body in a non-intrusive setting, by calibrating physiological arousal and engagement from face videos leading towards a hidden state of self-efficacy.