Jul 2026· Journal of imaging informatics in medicine· 0 citations· 43 references
Medicine
TL;DR
Overall, while RAG shows promise for improving factual grounding in radiology AI, current evaluation paradigms likely overestimate real-world clinical readiness and future work should prioritize retrieval quality, clinically grounded evaluation, safety-critical error analysis, bias assessment, and deployment-relevant efficiency metrics to enable responsible clinical translation.
Automatic radiology report generation has gained increasing attention for its potential to assist in clinical reporting and reduce the workload of radiologists. Existing 3D radiology report generation methods employ multi-modal foundation model to encode volume-text inputs and produce diagnosis reports, while they ignore the characteristics of 3D volumes including much background regions and suffer from generating hallucinations, especially in medical domain that contains many uncommon professional terms. In this paper, we aim to efficiently adapt the pre-trained foundation model to specific 3D radiology report generation, and present a Prior-drIven Prompting with diagnosis-oriented retrieval-Augmentation (PIPA) framework. In PIPA, we design a Prior-drIven Prompting (PIP) strategy to exploit diagnostic knowledge from input volumes and a Diagnosis-oriented volume-report retrievalaugmentation Generation (DIG) module to explore beneficial knowledge from external database. Specifically, in PIP, to take full advantage of the patient's clinical information, e.g., age and symptoms, and the possible disease information, e.g., brain tumor, edema, we formulate them as the patient and disease priors to mine clinical relevant knowledge. Furthermore, we propose utilizing visual and textual embeddings as queries to retrieve similar external data by devising a diagnosis-oriented retrieval-augmentation scheme for leveraging more report resources as references for LLM to produce accuracy outcomes. With PIP and DIG, PIPA integrates clinical priors and external data to learn effective diagnostic representations for high-quality report generation. We evaluate the framework on both public and in-house 3D medical datasets with corresponding reports, demonstrating its strong performance in generating accurate diagnosis reports. Source codes have been published at https://github.com/CUHK-AIM-Group/PIPA/tree/ main.
Qiushi Yang, Wuyang Li, Xiaoqing Guo et al.· IEEE Transactions on Medical...· 0 citations
Vision foundation models (VFMs) are increasingly being developed for radiological imaging, yet their definition, development and evaluation remain heterogeneous. We conducted a PRISMAScR scoping review of peer-reviewed studies published between January 2017 and March 2026 describing foundation models trained exclusively on radiological imaging data. Sixty-seven studies were included and mapped across three pillars: data scale and heterogeneity, architectural and pretraining scalability, and downstream transferability and generalization. Datasets primarily covered brain MRI, thoracoabdominal CT, and chest X-ray, ranging from fewer than 100,000 samples to multi-million-image cohorts. Transformer-based architectures and self-supervised pretraining predominated, particularly masked image modeling, contrastive learning and multi-stage approaches. Evaluation focused mainly on segmentation and classification, whereas cross-center, cross-scanner, anatomical and modality-shift validation was inconsistently reported. Alignment with FUTURE-AI principles was uneven. Overall, radiology-specific VFMs show promising transferability, but clinical translation remains constrained by limited data representativeness, heterogeneous benchmarks, incomplete reporting and insufficient deployment-oriented evaluation.
A. Vergara-Richart, Xavier Rafael-Palou, A. Fuster-Matanzo et al.· 0 citations
An evidence-guided retrieval-augmented framework EviRAG is proposed that decomposes retrieval into structured and unstructured alignment levels and substantially reduces clinical inconsistencies compared to strong medical vision-language baselines.
Yiyang Gu, Jiayue Fan, Kaili-May Liu et al.· Annual International ACM SIG...· 1 citation
Harrison.Rad 1.5 is the only system evaluated to meet the simulated FRCR passing standard and achieves the highest accuracy on closed-format clinical questions, across anatomical regions, on internal multi-body-part and mammography reporting, and on the primary clinically-aligned score for public chest reporting.
Suneeta Mall, V. Nekrasov, Ashnil Kumar et al.· 0 citations
BACKGROUND
Large language models (LLMs) show promise for converting complex radiology reports into patient-centric language, but inherent output instability may limit clinical application.
OBJECTIVES
To quantitatively assess the translational accuracy, error rates, and instability of various LLMs when generating patient-centric radiology reports, and evaluate demographic influences on report readability.
MATERIALS AND METHODS
This retrospective study evaluated 320 de-identified radiology reports processed by three LLMs using a two-stage (baseline and optimized) prompt engineering strategy. Two senior radiologists evaluated medical accuracy, completeness, and recommendation suitability. Readability was evaluated by 16 non-medical participants stratified by age and education.
RESULTS
Professional radiological evaluation revealed that all tested models exhibited inherent instability, omitted information, and tended to generate risk-averse, generalized clinical recommendations. To address these limitations, optimized structured prompts significantly reduced model output variance and improved translational accuracy, with particularly prominent effects observed in DeepSeek-R1 and ChatGPT-4.0. Overall, large language models significantly enhanced the readability of radiology reports (P < 0.05), with DeepSeek-R1 achieving the best performance. However, patients' self-reported comprehension of the reports was affected by demographic characteristics.
CONCLUSION
Large language models can effectively improve the readability of radiology reports, yet all such models inherently suffer from output instability and information omission. Optimized structured prompting can substantially reduce the variability of model outputs and improve the accuracy of medical text translation. Nevertheless, LLMs should currently be strictly confined to human-supervised auxiliary tools rather than applied as standalone clinical solutions.
Yun Mao, Chunyan Wang, Wei Wang et al.· Academic Radiology· 0 citations
Vision-language models (VLMs) are increasingly being evaluated for medical imaging, but many available benchmarks emphasize disease classification, report generation, or broad visual question answering rather than the spatial and anatomical reasoning required for radiology. We developed the Spatial Perception and Anatomical Reasoning in Clinical Radiology (SPARC-Rad) Benchmark, a manually curated multimodal benchmark dataset and evaluation pipeline for assessing these capabilities in radiology VLMs. SPARC-Rad includes 300 image-question pairs derived from healthy control imaging studies in The Cancer Imaging Archive (TCIA), spanning CT, MRI, and radiography across the abdomen, chest, breast, neuro, and musculoskeletal categories. Radiology trainees manually designed and annotated questions to evaluate anatomical identification, localization, laterality, regional recognition, device identification, and inter-structure spatial relationships. The evaluation pipeline supports standardized prompting, structured output collection, response normalization, LLM-as-judge grading, human quality review, binary correctness scoring, and subgroup analysis by modality, anatomy, and reasoning type. SPARC-Rad provides a reusable framework for evaluating whether VLMs can provide reasoning for radiologic anatomy as a spatial system, supporting future model development, failure-mode analysis, and pre-deployment assessment.
Satvik Tripathi, M. E. Seker, K. Quevada et al.· 0 citations
