IACTA-EST 2023 Winning Teams
Congratulations to the two winning teams for the 2023 edition of the IACTA-EST Challenge!
Artificial Doctors
Michael Yeung
Department of Computing, Imperial College London, London, UK; Lucida Medical Ltd, University of Cambridge Biomedical Innovation Hub, Cambridge, UK
AI Cracker 2023
Yanbo J. Wang, Sheng Chen, Hui Qin, Zhe Sun, Na Zhang, Min Yuan, Xuan Yang
LYZD-FinTech Co., LTD , Beijing, China
Qu Liu, Yan Li
The First Medical Center, Chinese PLA General Hospital, Beijing, China
2023 Leaderboard
Task 1: LVO Detection
As a reference, here are the scores of a bootstrapped random classifier 95% confidence intervals:
- Area under the ROC curve (AuROC): 0.38-0.62
- Average Precision Score (mAp): 0.41-0.61
- Total score (AuROC and mAp average): 0.41-0.61
Artificial Doctors and AI Cracker 2023 teams reached essentially equivalent predictive performance, and they were well above the random classifier confidence intervals.
Task 2: Reperfusion Prediction
As a reference, here are the scores of a bootstrapped random classifier 95% confidence intervals:
- Area under the ROC curve (AuROC): 0.27-0.73
- Average Precision Score (mAp): 0.27-0.66
- Total score (AuROC and mAp average): 0.27-0.73
No team was above the random classifier confidence intervals. Therefore, all teams have equally low predictive performance