TY - GEN
T1 - Enhancing AI-Assisted Stroke Emergency Triage with Adaptive Uncertainty Estimation
AU - Yang, Shuhua
AU - Cai, Tongan
AU - Ni, Haomiao
AU - Ma, Wenchao
AU - Xue, Yuan
AU - Wong, Kelvin
AU - Volpi, John
AU - Wang, James Z.
AU - Huang, Sharon X.
AU - Wong, Stephen T.C.
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2025/9/25
Y1 - 2025/9/25
N2 - Stroke diagnosis in emergency rooms (ERs) is challenging due to limited access to MRI scans and a shortage of neurologists. Although AI-assisted triage has shown promise, existing methods typically use MRI-derived training labels, which may not align with stroke patterns in patient multimedia data. To address this mismatch, we propose an Adaptive Uncertainty-aware Stroke TrIage Network (AUSTIN) (Source code for the framework is at https://github.com/shuashua0608/AUSTIN), that leverages inconsistencies between clinician triage decisions and MRI-derived labels to enhance AI-driven stroke triage. This approach mitigates overfitting to clinician-MRI disagreement cases during training, significantly improving test accuracy. Additionally, it identifies high-uncertainty samples during inference, prompting further imaging or expert review. Evaluated on a clinical stroke patient dataset collected in an ER setting, AUSTIN achieves over 20% performance gain over human triage and a 13% improvement over a prior state-of-the-art method. The learned uncertainty scores also show strong alignment with discrepancies in clinical assessments, highlighting the framework’s potential to enhance the reliability of AI-assisted stroke triage.
AB - Stroke diagnosis in emergency rooms (ERs) is challenging due to limited access to MRI scans and a shortage of neurologists. Although AI-assisted triage has shown promise, existing methods typically use MRI-derived training labels, which may not align with stroke patterns in patient multimedia data. To address this mismatch, we propose an Adaptive Uncertainty-aware Stroke TrIage Network (AUSTIN) (Source code for the framework is at https://github.com/shuashua0608/AUSTIN), that leverages inconsistencies between clinician triage decisions and MRI-derived labels to enhance AI-driven stroke triage. This approach mitigates overfitting to clinician-MRI disagreement cases during training, significantly improving test accuracy. Additionally, it identifies high-uncertainty samples during inference, prompting further imaging or expert review. Evaluated on a clinical stroke patient dataset collected in an ER setting, AUSTIN achieves over 20% performance gain over human triage and a 13% improvement over a prior state-of-the-art method. The learned uncertainty scores also show strong alignment with discrepancies in clinical assessments, highlighting the framework’s potential to enhance the reliability of AI-assisted stroke triage.
KW - Multimedia
KW - Stroke Triage
KW - Uncertainty Estimation
UR - https://www.scopus.com/pages/publications/105018059408
UR - https://www.scopus.com/inward/citedby.url?scp=105018059408&partnerID=8YFLogxK
U2 - 10.1007/978-3-032-05185-1_15
DO - 10.1007/978-3-032-05185-1_15
M3 - Conference contribution
AN - SCOPUS:105018059408
SN - 9783032051844
T3 - Lecture Notes in Computer Science
SP - 148
EP - 157
BT - Medical Image Computing and Computer Assisted Intervention, MICCAI 2025 - 28th International Conference, 2025, Proceedings
A2 - Gee, James C.
A2 - Hong, Jaesung
A2 - Sudre, Carole H.
A2 - Golland, Polina
A2 - Park, Jinah
A2 - Alexander, Daniel C.
A2 - Iglesias, Juan Eugenio
A2 - Venkataraman, Archana
A2 - Kim, Jong Hyo
PB - Springer Science and Business Media Deutschland GmbH
T2 - 28th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025
Y2 - 23 September 2025 through 27 September 2025
ER -