A Large Language Model Framework for Patient-Centered Translation of Radiology Reports Through Semantic Simplification and Health Literacy Enhancement
DOI:
https://doi.org/10.68339/Keywords:
Large Language Models (LLMs); Radiology Report Simplification; Patient-Centered Communication; Health Literacy; Chest X-ray Reports.Abstract
The language of radiology reports is often difficult to understand and can impede communication and/or informed clinical decisions. This research investigates the use of large language models (LLM) to create a patient-centred report from radiologist-generated chest X-ray reports without compromising clinical accuracy. The framework leverages the Indiana University OpenI Chest X-ray dataset, which contains about 4,000 radiology reports, and tests several state-of-the-art LLMs, GPT-4, Llama 3, Mistral, and Gemma. Radiology reports are converted to plain language summaries with enhanced health literacy but without affecting the diagnostic significance. The metrics employed to evaluate the performance are semantic similarity scores (BERTScore, BLEU, ROUGE-L), readability scores (Flesch Reading Ease, Flesch-Kincaid Grade Level), and comparison analysis between different LLM models. The proposed framework offers a systematic evaluation to assess the potential of contemporary LLMs in creating clinically accurate, patient-centric, and comprehensible radiology reports, ensuring transparency in healthcare communication and elevating patient engagement.
References
[1] Y. Yao, Z. Wen, Y. Tong, X. Tian, X. Li, X. Ma, D. Xu, and T. Gedeon, “Thought graph traversal for test-time scaling in chest X-ray VLLMs,” Pattern Recognition, vol. 179, 2026, doi: 10.1016/j.patcog.2026.113639.
[2] J. Park, S. Kim, B. Yoon, J. Hyun, and K. Choi, “M4CXR: Exploring multitask potentials of multimodal large language models for chest X-ray interpretation,” IEEE Transactions on Neural Networks and Learning Systems, vol. 36, no. 10, 2025, doi: 10.1109/TNNLS.2025.3587687.
[3] J. Xu, “Uncertainty estimation in large vision language models for automated radiology report generation,” in Proc. Machine Learning for Health (ML4H), vol. 259, 2024.
[4] C. Pellegrini, E. Özsoy, F. T. Gassert, A. W. Marka, M. Strenzke, M. Keicher, M. R. Makowski, and N. Navab, “Enhancing radiology workflows through collaborative AI-assisted chest X-ray reporting using large vision-language models: A proof-of-concept study,” Insights into Imaging, vol. 17, no. 1, 2026, doi: 10.1186/s13244-026-02292-7.
[5] W.-J. Noh, S.-W. Pi, and B.-D. Lee, “Hybrid framework for lesion-aware, clinically coherent chest X-ray report generation using contrastive learning and large language models,” Scientific Reports, vol. 16, no. 1, 2026, doi: 10.1038/s41598-025-34799-2.
[6] R. B. Lanfredi, P. Mukherjee, and R. M. Summers, “Enhancing chest X-ray datasets with privacy-preserving large language models and multi-type annotations: A data-driven approach for improved classification,” Medical Image Analysis, vol. 99, 2025, doi: 10.1016/j.media.2024.103383.
[7] X. Mei, L. Yang, D. Gao, X. Cai, J. Han, and T. Liu, “FiR-Rad: Fine-grained reinforcement with structured reasoning for chest X-ray report generation,” IEEE Transactions on Medical Imaging, vol. 45, no. 6, 2026, doi: 10.1109/TMI.2026.3665561.
[8] M. Barzegar, H. Rostami, A. Sanati, R. Afshoon, A. Kosari, and A. Keshavarz, “Enhancing chest X-ray report generation with pathology-guided prompts and vision-language shortcut bias mitigation,” Biomedical Signal Processing and Control, vol. 120, 2026, doi: 10.1016/j.bspc.2026.110173.
[9] R. W. Lee, K. H. Lee, J. S. Yun, M. S. Kim, and H. S. Choi, “Comparative analysis of M4CXR, an LLM-based chest X-ray report generation model, and ChatGPT in radiological interpretation,” Journal of Clinical Medicine, vol. 13, no. 23, 2024, doi: 10.3390/jcm13237057.
[10] G. A. P. Rodrigues, A. L. M. Serrano, G. D. Bispo, G. P. Rocha, V. P. Gonçalves, and R. I. Meneguette, “IHRAS: Automated medical report generation from chest X-rays via classification, segmentation, and LLMs,” Bioengineering, vol. 12, no. 8, 2025, doi: 10.3390/bioengineering12080795.
[11] H. Ali, R. I. Sumon, A. R. Khalid, K. Fathima, and H. C. Kim, “A semantic evaluation framework for medical report generation using large language models,” CMC-Computers, Materials & Continua, vol. 84, no. 3, 2025, doi: 10.32604/cmc.2025.065992.
[12] S. Allaberdiev, A. Khan, S. Mamarasulov, and X. Chen, “ChestXGen: Dynamic memory-augmented vision-language transformer with context-aware gating for radiology report generation,” Journal of Artificial Intelligence and Soft Computing Research, vol. 16, no. 1, 2026, doi: 10.2478/jaiscr-2026-0003.
[13] S. Wang, Z. Zhao, X. Ouyang, T. Liu, Q. Wang, and D. Shen, “Interactive computer-aided diagnosis on medical image using large language models,” Communications Engineering, vol. 3, no. 1, 2024, doi: 10.1038/s44172-024-00271-8.
[14] Y. Tian and Y. Song, “Feature decomposition via shared low-rank matrix recovery for CT report generation,” IEEE Transactions on Medical Imaging, vol. 45, no. 4, 2026, doi: 10.1109/TMI.2025.3628159.
[15] Y. Liu, Y. Li, Z. Wang, X. Liang, L. Liu, L. Wang, L. Cui, Z. Tu, L. Wang, and L. Zhou, “A systematic evaluation of GPT-4V’s multimodal capability for chest X-ray image analysis,” MetaRadiology, vol. 2, no. 4, 2024, doi: 10.1016/j.metrad.2024.100099.
[16] S. Ju, M. Wang, Y. Xiao, Y. Shen, J. Cheng, Y. Su, and X. Liao, “DM-RRG: Radiology image report generation with disease-topic guidance and multi-grained image-text alignment,” Biomedical Signal Processing and Control, vol. 114, 2026, doi: 10.1016/j.bspc.2025.109322.
[17] L. Yang, X. Liang, Z. Wang, Z. Diao, X. Huang, D. Shen, X. Tan, H. Li, Z. Chen, S. Qiu, and L. Zhou, “Constructing a unified vision-language model for chest radiograph-based diagnostics, medical education, and data augmentation,” Radiology: Cardiothoracic Imaging, vol. 7, no. 6, 2025, doi: 10.1148/ryct.250033.
[18] A. Izhar, N. Idris, and N. Japar, “Engaging preference optimization alignment in large language model for continual radiology report generation: A hybrid approach,” Cognitive Computation, vol. 17, no. 1, 2025, doi: 10.1007/s12559-025-10404-6.
[19] S. Tripathi, L. Mutter, M. Muppuri, S. Dheer, E. Garza-Frias, K. Awan, A. Jha, M. Dezube, A. Tabari, B. C. Bizzo, K. J. Dreyer, C. P. Bridge, and D. Daye, “PRECISE framework: Enhanced radiology reporting with GPT for improved readability, reliability, and patient-centered care,” European Journal of Radiology, vol. 187, 2025, doi: 10.1016/j.ejrad.2025.112124.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Muhammad Ahsan (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
