Affective Computing in Anxiety Disorders: A Rapid Literature Review of Emotion Recognition Applications

Le résumé fourni par la source

Anxiety disorders (ADs) affect roughly one in ten people in the UK, and this number is expected to increase, intensifying the need for innovation. Digital technologies such as affective computing (AC, technology to detect human emotions) could foster a more patient-centric approach, enhancing therapy adherence and optimizing clinician-patient interactions. This paper reviews the literature relevant to the integration of affective computing in clinical pathways for ADs. A search was conducted on Google Scholar and PubMed using the keywords “affective computing” and subtypes of anxiety disorders. A total of 355 results were filtered to focus on peer-reviewed articles that specifically addressed emotion recognition in pathological anxiety as opposed to simply feeling anxious. Findings underscore prevalent studies focusing on post-traumatic stress disorder (PTSD) and the widespread use of valence and arousal for emotion quantification. Various approaches for both eliciting and detecting emotions are explored, offering technical and practical insights. Diverse applications, from monitoring treatment progression in behavioral therapies to assessing the efficiency of deep brain stimulation for intractable obsessive-compulsive disorder, highlight affective computing's versatility and promise. A significant advantage of digital technologies is their potential to capture longitudinal and contextualized data beyond clinical confines. Such assessments elucidate patients' daily challenges and triggers, enabling tailored interventions. The literature suggests that AC has the potential to support mental healthcare and improve patient outcomes. However, further evidence of its effective benefits is required, especially for ADs beyond PTSD, and further exploration of its implementation in clinical pathways is needed.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.