Digital Phenotyping as Felt Informatics: Designing AI-Based Mental Health Diagnostic Tools Through Aesthetics

要旨

With psychiatry lagging behind other medical fields in terms of innovation in instruments and methods, AI provides it an opportunity to catch up. Advocates of digital phenotyping promise to provide an objective tool that detects symptoms by analysing data from personal devices. We argue that digital phenotyping requires a more reflexive and critical approach to its design and an alignment of the clinicians' interests in generating relevant evidence with the needs of service users who seek tools to manage their condition. We propose a felt informatics approach, situating digital phenotyping design within the problem space of pragmatist aesthetics. Within this perspective, felt life becomes a central object and a site for digital phenotyping design. This paper reveals the ways diagnostic data mediates mental ill health experience, emphasises the cultivation of aesthetic sensibility as a fundamental element of digital phenotyping and includes design considerations for practitioners and researchers.

著者
Karin Bogdanova
TU Delft, Delft, Netherlands
Nazli Cila
TU Delft, Delft, Netherlands
Olya Kudina
TU Delft, Delft, Netherlands
Alessandro Bozzon
TU Delft, Delft, Netherlands
DOI

10.1145/3706598.3714399

論文URL

https://dl.acm.org/doi/10.1145/3706598.3714399

会議: CHI 2025

The ACM CHI Conference on Human Factors in Computing Systems (https://chi2025.acm.org/)

セッション: Health and Well-being

G303
7 件の発表
2025-04-30 01:20:00
2025-04-30 02:50:00
日本語まとめ
読み込み中…