Radiology report generation adapts to varying clinical data availability
PACER: Progressive Availability-Conditioned Evidence Routing for Radiology Report Generation under Incomplete Clinical Context
Computer Vision and Pattern Recognition
Summary
Radiology reports often rely on different kinds of medical information, but not all information is always available for each patient. This paper presents a new method called PACER that progressively uses whatever evidence is present to make more accurate and reliable reports. The system refines image data, adjusts text generation based on what clues it has, and creates structured clinical statements to guide the report. This approach improves the clinical accuracy of reports even when some patient information is missing.
What this means in practice
- •For hospital radiology teams: Generate more clinically accurate radiology reports when some patient information is incomplete or missing during imaging exams.
- •For medical ai developers: Incorporate progressive evidence routing methods into AI tools that create radiology reports conditioned on variable available clinical data.
Authors
Yulong Chen, Yadong Liu, Haoyu Cao, Sen Xu, Yueying Wang, Jie Wen
Abstract
Radiology report generation (RRG) increasingly incorporates heterogeneous clinical evidence, such as multi-view radiographs and previous reports, whose availability varies across examinations. However, accommodating different input combinations does not ensure effective evidence use: generated reports may still omit or inaccurately describe clinically relevant findings. To address this problem, we propose PACER, a Progressive Availability-Conditioned Evidence Routing framework for structured incomplete-context RRG that follows a Refine-Calibrate-Commit pipeline. It first refines observed visual representations through endpoint-preserving patchwise routing across frozen encoder depths, incorporating complementary cues while retaining the pretrained terminal representation. It then calibrates the language-model prefix according to the observed evidence and availability state, adapting the shared generator's conditioning as the available source set changes. Finally, it generates polarity-structured clinical commitments before the report in the same autoregressive trajectory, providing structured clinical context for subsequent generation. Experiments demonstrate state-of-the-art clinical efficacy across all four MIMIC-RG4 settings and strong MIMIC-CXR performance, while maintaining competitive language-generation quality.