Paper Number
ICIS2025-2158
Paper Type
Short
Abstract
Mental health assessments increasingly rely on remote formats that produce rich but complex multimodal data in the form of text, audio, and video, processed through multiple machine learning models. While these environments offer new opportunities for insight, they also pose significant challenges for effective clinical sensemaking. This study introduces a dashboard designed to address these challenges by enabling practitioners to explore behavioral data from multiple angles while mitigating overreliance on model accuracy metrics. Grounded in the Design Science Research, the dashboard design is informed by an integration of Integrative Sensemaking Theory and Signal Detection Theory. The research contributes a set of design requirements for supporting sensemaking in multimodal, multi-model contexts, instantiates them in a dashboard artifact, and proposes an evaluation with practitioners using clinically grounded sensemaking tasks. This work advances computational design by offering theoretical and practical insights for advancing the integration of these complex data in mental healthcare context.
Recommended Citation
Ademaj, Gemza; Zhang, Xinyuan; Abbasi, Ahmed; Sarker, Saonee; and Sarker, Suprateek, "Designing Support for Sensemaking in Multimodal, Multi-model Mental Health Assessments" (2025). ICIS 2025 Proceedings. 15.
https://aisel.aisnet.org/icis2025/is_health/ishealthcare/15
Designing Support for Sensemaking in Multimodal, Multi-model Mental Health Assessments
Mental health assessments increasingly rely on remote formats that produce rich but complex multimodal data in the form of text, audio, and video, processed through multiple machine learning models. While these environments offer new opportunities for insight, they also pose significant challenges for effective clinical sensemaking. This study introduces a dashboard designed to address these challenges by enabling practitioners to explore behavioral data from multiple angles while mitigating overreliance on model accuracy metrics. Grounded in the Design Science Research, the dashboard design is informed by an integration of Integrative Sensemaking Theory and Signal Detection Theory. The research contributes a set of design requirements for supporting sensemaking in multimodal, multi-model contexts, instantiates them in a dashboard artifact, and proposes an evaluation with practitioners using clinically grounded sensemaking tasks. This work advances computational design by offering theoretical and practical insights for advancing the integration of these complex data in mental healthcare context.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
21-Healthcare