Back

CLEAR: An Auditable Foundation Model for Radiology Grounded in Clinical Concepts

Han, T.; Wu, R.; Tian, Y.; Khader, F.; Adams, L.; Bressem, K.; Davatzikos, C.; Kather, J. N.; Shen, L.; Mankoff, D.; Barbosa, E.; Truhn, D.

2026-01-17 health informatics
10.64898/2026.01.15.26344222 medRxiv
Show abstract

"Black box" deep learning models for medical image interpretation limit clinical trust and analysis of performance degradation. Here, we introduce Concept-Level Embeddings for Auditable Radiology (CLEAR), an auditable foundation model based on clinical concepts. Trained on over 0.87 million image-report pairs from 239,091 patients, CLEAR learns a visual representation and projects chest X-rays into a semantically rich space defined by large language model embeddings, making every prediction traceable to specific radiological observations. External validation on four large, physician-annotated datasets from the United States, Europe, and Asia shows that CLEAR not only achieves state-of-the-art classification performance but also enables novel applications: auditable zero-shot pathology detection, systematic identification of radiological confounders, and the creation of expert-level concept bottleneck models from data-driven concepts. By integrating clinical knowledge directly into its reasoning process, CLEAR offers a framework for robust model auditing, safer deployment, and enhanced physician-AI collaboration, advancing towards trustworthy medical AI.

Published in Nature Biomedical Engineering (predicted rank #20) · training set

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.