ScholarDeck

AI / informaticsmedRxiv : the preprint server for health sciences · 2026

A framework for human-artificial intelligence co-learning for disease activity labeling using electronic health records.

Zongxin Yang, Yuming Zhang, Zoe Love … Katherine P Liao

RheumatologyAI / informatics

Human-AI co-learning reduces cost for disease activity labeling in electronic health records.

A framework for human-artificial intelligence co-learning for disease activity labeling using electronic health records.

Zongxin Yang … Katherine P Liao

medRxiv : the preprint server for health sciences · 2026

Purpose

To develop and evaluate a framework for human-AI interaction.

Methods

Using rheumatoid arthritis (rheumatoid arthritis) disease activity as the use- case, we studied a multi-institutional electronic health record-based rheumatoid arthritis cohort of 3,167 patients.

n = 3,167 patients

69%

Results

69%

reduced the estimated cost of identifying disease activity in patient records

n = 3,167 patients

More results

Expert reviewers labeled 626 notes from 273 patients; human-AI adjudication revised 127 (20%) of these initial labels and added 60 newly labeled notes, yielding a 686-note co-produced reference.

More results

Against this reference, the final agent's accuracy improved from a mean absolute error of 0.406 to 0.291 with co-learning, and its ambiguity flag agreed with expert ambiguity designations with 92.1% accuracy.

“

Conclusion · 1 of 2

Adopting a framework for human-AI co-learning, SHARE, improved the overall quality of gold-standard labels, identified ambiguous cases for further review, and supported accurate and standardized chart reviews of disease activity at a scale infeasible for manual review.

Conclusion · 2 of 2

SHARE's resource efficiency provides a transferable approach to incorporate complex phenotypes in real-world evidence studies.

Post

A framework for human-artificial intelligence co-learning for disease activity labeling using electronic health records.

Human-AI co-learning reduces cost for disease activity labeling in electronic health records.

Related papers