Multimodal Fusion of Linguistic, Physiological, and Behavioral Fatigue Signals

Integrate linguistic anomaly detection with contemporaneous physiological and behavioral fatigue signals from wearables and smartphones (e.g., sleep debt, circadian disruption), along with EHR metadata such as overnight shifts and patient volume. This multimodal approach aims to triangulate fatigue states more accurately and reduce bias from relying on a single modality, enhancing the robustness of fatigue detection and its relationship to clinical decision-making.

References:

1. Impact of Fatigue on Emergency Physicians’ Decision-making for Computed Tomographic Scan Requests and Inpatient Referrals: An Observational Study from a Tertiary Care Medical Center of the Sultanate of Oman. Asma H Al-Arimi, D. Hazra, A. Al-Alawi (2023). Indian Journal of Critical Care Medicine.
2. Clinical Notes Reveal Physician Fatigue. Chao-Chun Hsu, Ziad Obermeyer, Chenhao Tan (2023). arXiv.org.
3. A machine learning model using clinical notes to identify physician fatigue. Chao-Chun Hsu, Ziad Obermeyer, Chenhao Tan (2025). Nature Communications.
4. Clinical Notes Reveal Physician Fatigue. Chao-Chun Hsu, Ziad Obermeyer, Chenhao Tan (2023). arXiv.org.
5. Prescribing decision making by medical residents on night shifts: A qualitative study. J. Lauffenburger, M. Coll, Erin Kim, Ted Robertson, Rebecca Oran, Nancy Haff, Kaitlin Hanken, J. Avorn, N. Choudhry (2022). Medical Education.
6. Development of prediction models for screening depression and anxiety using smartphone and wearable-based digital phenotyping: protocol for the Smartphone and Wearable Assessment for Real-Time Screening of Depression and Anxiety (SWARTS-DA) observational study in Korea. Yu-Bin Shin, Ah Young Kim, Seonmin Kim, Min-Sup Shin, Jinhwa Choi, Kyung Lyun Lee, Jisu Lee, Sangwon Byun, Sujin Kim, Heon-Jeong Lee, Chul-Hyun Cho (2025). BMJ Open.