Back to AI News
arxiv

ARIVE interleaves anatomical inspection and diagnostic reasoning for explainable radiology reports

Researchers introduce ARIVE, an explainable radiology report generation framework that interleaves anatomical inspection with diagnostic reasoning before composing the final report. They constructed 159,416 multimodal chain-of-thought trajectories from MIMIC-CXR linking inspected anatomical regions to clinical interpretation, then trained and optimized the model with anatomical and clinical-content rewards. ARIVE showed superior clinical-efficacy performance versus strong baselines on MIMIC-CXR and MIMIC-ABN, and its reports were competitive with leading models in blinded expert grading. The trajectories earned high diagnostic-traceability ratings, and a paired review evaluation suggests they can help identify errors in generated reports.

AI Brief

It presents a concrete, evaluated method for making AI-generated radiology reports more traceable and reviewable, which matters for clinical trust and error correction.

arXiv:2611.14018v1 Announce Type: cross Abstract: Automated radiology report generation can draft clinically plausible reports, yet the pathway from visual evidence to reported findings remains opaque, leaving the diagnostic basis of these reports unexplained and the generated reports difficult to review or revise. We introduce ARIVE (Anatomy-guided Reasoning with Interleaved Visual Evidence), an explainable report generation framework that interleaves anatomical inspection with diagnostic reasoning prior to final report composition. To support this process, we constructed 159,416 multimodal chain-of-thought (MCoT) trajectories from MIMIC-CXR, wherein visual evidence from inspected anatomical regions informs subsequent clinical interpretation. Trained on these trajectories and further optimized using anatomical and clinical-content rewards, ARIVE achieved superior clinical-efficacy performance compared with strong baselines on MIMIC-CXR and the abnormality-centered MIMIC-ABN. ARIVE's reports were competitive with those of leading models in blinded expert grading, and its MCoT trajectories earned high ratings for diagnostic traceability. Furthermore, a paired review evaluation suggests that access to MCoT can aid in identifying errors within generated reports.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!