大语言模型GPT在医疗文本中的应用综述

doi:10.3778/j.issn.1673-9418.2409076

摘要/Abstract

摘要： 医疗文本是医学知识记录与传递的重要载体，但随着医疗数据的迅猛增长，传统的人工处理方式已难以满足日益增长的效率与准确性需求。近年来，以GPT为代表的大语言模型在自然语言处理领域取得突破，具备强大的语言理解与生成能力，为高效处理医疗文本提供了新思路。介绍了大语言模型GPT的核心技术原理，并重点分析其在医疗数据处理、辅助医患沟通、医学教育支持、疾病预防管理以及多模态综合应用等五大领域中的实际应用；系统总结了大语言模型GPT在医疗文本处理方面所展现出的信息整合效率高、医学知识储备丰富等方面的优势；深入探讨了其在实际应用中暴露的问题，并给出了具有可行性的解决思路与技术优化方向；结合当前技术发展趋势，展望了大语言模型在医疗领域的未来应用前景。

关键词: 大语言模型, 医疗文本, GPT, 自然语言处理

Abstract: Medical texts serve as a crucial medium for recording and conveying medical knowledge. However, with the rapid growth of medical data, traditional manual processing methods are increasingly unable to meet the demands for efficiency and accuracy. In recent years, large language models, represented by GPT (generative pre-trained transformer), have made significant breakthroughs in the field of natural language processing. With their powerful language understanding and generation capabilities, they offer new approaches for the efficient processing of medical texts. This paper introduces the core technical principles of GPT and focuses on its practical applications in five key areas: medical data processing, doctor-patient communication assistance, medical education support, disease prevention and management, and multimodal integration. It systematically summarizes the advantages GPT exhibits in medical text processing, particularly its high efficiency in information integration and rich medical knowledge base. Furthermore, this paper explores the challenges exposed in real-world applications, proposing feasible solutions and directions for technical optimization. Based on current technological trends, it also envisions the prospects of large language models in healthcare domain.

Key words: large language models, medical text, GPT, natural language processing

田崇腾, 刘静, 王晓燕, 李明. 大语言模型GPT在医疗文本中的应用综述[J]. 计算机科学与探索, 2025, 19(8): 2043-2056.

TIAN Chongteng, LIU Jing, WANG Xiaoyan, LI Ming. Review of Application of Large Language Models GPT in Medical Text[J]. Journal of Frontiers of Computer Science and Technology, 2025, 19(8): 2043-2056.

参考文献

[1] MOOR M, BANERJEE O, ABAD Z S H, et al. Foundation models for generalist medical artificial intelligence[J]. Nature, 2023, 616(7956): 259-265.
[2] CHOWDHARY K R. Natural language processing[M]// Fundamentals of artificial intelligence. New Delhi: Springer India, 2020: 603-649.
[3] THIRUNAVUKARASU A J, TING D S J, ELANGOVAN K, et al. Large language models in medicine[J]. Nature Medicine, 2023, 29(8): 1930-1940.
[4] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.
[5] SAMEK W, MONTAVON G, LAPUSCHKIN S, et al. Explaining deep neural networks and beyond: a review of methods and applications[J]. Proceedings of the IEEE, 2021, 109(3): 247-278.
[6] ZENG W, REN X Z, SU T, et al. PanGu-α: large-scale autoregressive pretrained Chinese language models with auto-parallel computation[EB/OL]. [2024-07-19]. https://arxiv.org/ abs/2104.12369.
[7] MANN D L. Artificial intelligence discusses the role of artificial intelligence in translational medicine: a JACC: basic to translational science interview with ChatGPT[J]. JACC Basic to Translational Science, 2023, 8(2): 221-223.
[8] TEAM G, ANIL R, BORGEAUD S, et al. Gemini: a family of highly capable multimodal models[EB/OL]. [2024-07-19]. https://arxiv.org/abs/2312.11805.
[9] KELLY D, CHEN Y M, CORNWELL S E, et al. Bing chat: the future of search engines?[J]. Proceedings of the Association for Information Science and Technology, 2023, 60(1): 1007-1009.
[10] 孔祥溢, 王任直. 人工智能及在医疗领域的应用[J]. 医学信息学杂志, 2016, 37(11): 2-5.
KONG X Y, WANG R Z. Artificial intelligence and its application in medical field[J]. Journal of Medical Informatics, 2016, 37(11): 2-5.
[11] HAUPT C E, MARKS M. AI-generated medical advice-GPT and beyond[J]. JAMA, 2023, 329(16): 1349-1350.
[12] LANDSCHAFT A, ANTWEILER D, MACKAY S, et al. Implementation and evaluation of an additional GPT-4-based reviewer in PRISMA-based medical systematic literature reviews[J]. International Journal of Medical Informatics, 2024, 189: 105531.
[13] WANG H Y, WU W Z, DOU Z, et al. Performance and exploration of ChatGPT in medical examination, records and education in Chinese: pave the way for medical AI[J]. International Journal of Medical Informatics, 2023, 177: 105173.
[14] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30, 2017: 5998-6008.
[15] BUBECK S, CHANDRASEKARAN V, ELDAN R, et al. Sparks of artificial general intelligence: early experiments with GPT-4[EB/OL]. [2024-07-19]. https://arxiv.org/abs/2303. 12712.
[16] CARVALKO J. GPT-a paradigm shift for the twenty-first century[EB/OL]. [2024-07-19]. https://www.techrxiv.org/doi/ full/10.36227/techrxiv.23690874.
[17] KNOX W B, STONE P. Augmenting reinforcement learning with human feedback[C]//Proceedings of the ICML Workshop on New Developments in Imitation Learning, 2011.
[18] DAI D M, SUN Y T, DONG L, et al. Why can GPT learn in-context? Language models implicitly perform gradient descent as meta-optimizers[EB/OL]. [2024-07-19]. https://arxiv.org/abs/2212.10559.
[19] KNOX W B, STONE P. Interactively shaping agents via human reinforcement: the TAMER framework[C]//Proceedings of the 5th International Conference on Knowledge Capture. New York: ACM, 2009: 9-16.
[20] LEIKE J, KRUEGER D, EVERITT T, et al. Scalable agent alignment via reward modeling: a research direction[EB/OL]. [2024-07-20]. https://arxiv.org/abs/1811.07871.
[21] SCHULMAN J, WOLSKI F, DHARIWAL P, et al. Proximal policy optimization algorithms[EB/OL]. [2024-07-20]. https:// arxiv.org/abs/1707.06347.
[22] YUAN D, RASTOGI E, NAIK G, et al. A continued pretrained LLM approach for automatic medical note generation[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2403.09057.
[23] GE W D, RICE H J, SHEIKH I S, et al. Improving neurology clinical care with natural language processing tools[J]. Neurology, 2023, 101(22): 1010-1018.
[24] SIRRIANNI J, SEZGIN E, CLAMAN D, et al. Medical text prediction and suggestion using generative pretrained transformer models with dental medical notes[J]. Methods of Information in Medicine, 2022, 61(5/6): 195-200.
[25] WAISBERG E, ONG J, MASALKHI M, et al. GPT-4: a new era of artificial intelligence in medicine[J]. Irish Journal of Medical Science (1971-), 2023, 192(6): 3197-3200.
[26] CHEN L, LI H, SU Y Q, et al. Using a google web search analysis to assess the utility of ChatGPT in stem cell therapy[J]. Stem Cells Translational Medicine, 2024, 13(1): 60-68.
[27] WU Z L, HASAN A, WU J G, et al. Chain-of-though (CoT) prompting strategies for medical error detection and correction[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2406.09103.
[28] GUAN Z H, WU Z H, LIU Z L, et al. CohortGPT: an enhanced GPT for participant recruitment in clinical study[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2307.11346.
[29] BALKUS S V, YAN D H. Improving short text classification with augmented data using GPT-3[J]. Natural Language Engineering, 2024, 30(5): 943-972.
[30] WADA A, AKASHI T, SHIH G, et al. Optimizing GPT-4 turbo diagnostic accuracy in neuroradiology through prompt engineering and confidence thresholds[J]. Diagnostics, 2024, 14(14): 1541.
[31] GIUFFRè M, MASUTTI F, GULOTTA M, et al. Optimizing AI performance for autoimmune hepatitis guidelines: a case study on enhanced GPT model accuracy[J]. Digestive and Liver Disease, 2024, 56: S28.
[32] SALVAGNO M, TACCONE F S, GERLI A G. Artificial intelligence hallucinations[J]. Critical Care, 2023, 27(1): 180.
[33] LEE P, BUBECK S, PETRO J. Benefits, limits, and risks of GPT-4 as an AI chatbot for medicine[J]. New England Journal of Medicine, 2023, 388(13): 1233-1239.
[34] CHELLI M, DESCAMPS J, LAVOUé V, et al. Hallucination rates and reference accuracy of ChatGPT and bard for systematic reviews: comparative analysis[J]. Journal of Medical Internet Research, 2024, 26: e53164.
[35] SOVRANO F, ASHLEY K, BACCHELLI A. Toward eliminating hallucinations: GPT-based explanatory AI for intelligent textbooks and documentation[C]//CEUR Workshop Proceedings, 2023: 54-65.
[36] SHAKIL H, ORTIZ Z, FORBES G C. Utilizing GPT to enhance text summarization: a strategy to minimize hallucinations[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2405. 04039.
[37] JAHANBAKHSH K, HAJIABADI M, GAGRANI V, et al. Beyond hallucination: building a reliable question answering & explanation system with GPTs[C]//Proceedings of the NeurIPS??23 Workshop on Generative AI for Education, 2023.
[38] AYERS J W, POLIAK A, DREDZE M, et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum[J]. JAMA Internal Medicine, 2023, 183(6): 589-596.
[39] GORDON E B, TOWBIN A J, WINGROVE P, et al. Enhancing patient communication with Chat-GPT in radiology: evaluating the efficacy and readability of answers to common imaging-related questions[J]. Journal of the American College of Radiology, 2024, 21(2): 353-359.
[40] LIU X, ZHENG Y, DU Z, et al. GPT understands, too[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2103.10385.
[41] LEE T C, STALLER K, BOTOMAN V, et al. ChatGPT answers common patient questions about colonoscopy[J]. Gastroenterology, 2023, 165(2): 509-511.
[42] LIU Z L, HUANG Y, YU X W, et al. DeID-GPT: zero-shot medical text de-identification by GPT-4[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2303.11032.
[43] ZENG C K, HE D B, FENG Q, et al. SecureGPT: a framework for multi-party privacy-preserving transformer inference in GPT[J]. IEEE Transactions on Information Forensics and Security, 2024, 19: 9480-9493.
[44] LEVKOVICH I, ELYOSEPH Z. Suicide risk assessments through the eyes of ChatGPT-3.5 versus ChatGPT-4: vignette study[J]. JMIR Mental Health, 2023, 10: e51232.
[45] HADAR-SHOVAL D, ELYOSEPH Z, LVOVSKY M. The plasticity of ChatGPT??s mentalizing abilities: personalization for personality structures[J]. Frontiers in Psychiatry, 2023, 14: 1234397.
[46] PARKER G, SPOELMA M J. A chat about bipolar disorder[J]. Bipolar Disorders, 2024, 26(3): 249.
[47] KUNG T H, CHEATHAM M, MEDENILLA A, et al. Performance of ChatGPT on USMLE: potential for AI-assisted medical education using large language models[J]. PLoS Digital Health, 2023, 2(2): e0000198.
[48] GILSON A, SAFRANEK C W, HUANG T, et al. How does ChatGPT perform on the United States medical licensing examination (USMLE)? The implications of large language models for medical education and knowledge assessment[J]. JMIR Medical Education, 2023, 9: e45312.
[49] JOHNSON D, GOODMAN R, PATRINELY J, et al. Assessing the accuracy and reliability of AI-generated medical responses: an evaluation of the Chat-GPT model[EB/OL]. [2024-07-20]. https://pmc.ncbi.nlm.nih.gov/articles/PMC10002821/.
[50] HUANG X, ESTAU D, LIU X, et al. Evaluating the performance of ChatGPT in clinical pharmacy: a comparative study of ChatGPT and clinical pharmacists[J]. British Journal of Clinical Pharmacology, 2024, 90(1): 232-238.
[51] SUáREZ A, DíAZ-FLORES GARCíA V, ALGAR J, et al. Unveiling the ChatGPT phenomenon: evaluating the consistency and accuracy of endodontic question answers[J]. International Endodontic Journal, 2024, 57(1): 108-113.
[52] LAI U H, WU K S, HSU T Y, et al. Evaluating the performance of ChatGPT-4 on the United Kingdom medical licensing assessment[J]. Frontiers in Medicine, 2023, 10: 1240915.
[53] TACK A, PIECH C. The AI teacher test: measuring the pedagogical ability of blender and GPT-3 in educational dialogues[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2205.07540.
[54] AMIN M Y M. AI and chat GPT in language teaching: enhancing EFL classroom support and transforming assessment techniques[J]. International Journal of Higher Education Pedagogies, 2023, 4(4): 1-15.
[55] WANG J D, YAO Z H, YANG Z C, et al. NoteChat: a dataset of synthetic doctor-patient conversations conditioned on clinical notes[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2310.15959.
[56] BEN ABACHA A, YIM W W, ADAMS G, et al. Overview of the MEDIQA-chat 2023 shared tasks on the summarization & generation of doctor-patient conversations[C]//Proceedings of the 5th Clinical Natural Language Processing Workshop. Stroudsburg: ACL, 2023: 503-513.
[57] RHODES G A, HUANG S H. Augmented and virtual reality in the world of GPT text and image creations: AI, metaverse, and art[M]//Augmented and virtual reality in the metaverse. Cham: Springer, 2024: 227-246.
[58] 袁方, 任海玲, 赵梦, 等. 我国传染病预警监测模型研究进展综述[J]. 价值工程, 2023, 42(33): 162-165.
YUAN F, REN H L, ZHAO M, et al. Review of research progress on early warning and monitoring models for infectious diseases in China[J]. Value Engineering, 2023, 42(33): 162-165.
[59] HASNAIN M. ChatGPT applications and challenges in controlling monkey pox in Pakistan[J]. Annals of Biomedical Engineering, 2023, 51(9): 1889-1891.
[60] HOWARD A, HOPE W, GERADA A. ChatGPT and antimicrobial advice: the end of the consulting infection doctor?[J]. The Lancet Infectious Diseases, 2023, 23(4): 405-406.
[61] PEREYRA L, SCHLOTTMANN F, STEINBERG L, et al. Colorectal cancer prevention: is chat generative pretrained transformer (Chat GPT) ready to assist physicians in determining appropriate screening and surveillance recommendations?[J]. Journal of Clinical Gastroenterology, 2024, 58(10): 1022-1027.
[62] ARSLAN S. Exploring the potential of chat GPT in personalized obesity treatment[J]. Annals of Biomedical Engineering, 2023, 51(9): 1887-1888.
[63] ZAHRA M A, AL-TAHER A, ALQUHAIDAN M, et al. The synergy of artificial intelligence and personalized medicine for the enhanced diagnosis, treatment, and prevention of disease[J]. Drug Metabolism and Personalized Therapy, 2024, 39(2): 47-58.
[64] PATRINOS G P, SARHANGI N, SARRAMI B, et al. Using ChatGPT to predict the future of personalized medicine[J]. The Pharmacogenomics Journal, 2023, 23(6): 178-184.
[65] WU C Y, LEI J Y, ZHENG Q Y, et al. Can GPT-4V (ision) serve medical applications? Case studies on GPT-4V for multimodal medical diagnosis[EB/OL]. [2024-07-19]. https:// arxiv.org/abs/2310.09909.
[66] YANG Z, YAO Z, TASMIN M, et al. Performance of multimodal GPT-4V on USMLE with image: potential for imaging diagnostic support with explanations[EB/OL]. [2024-07-19]. https://www.medrxiv.org/content/10.1101/2023.10.26.23297629v3.
[67] BRIN D, SORIN V, BARASH Y, et al. Assessing GPT-4 multimodal performance in radiological image analysis[J]. European Radiology, 2025, 35(4): 1959-1965.
[68] ZAMAN K T, HASAN W U, LI J, et al. Empowering caregivers of Alzheimer??s disease and related dementias (ADRD) with a GPT-powered voice assistant: leveraging peer insights from social media[C]//Proceedings of the 2023 IEEE Symposium on Computers and Communications. Piscataway: IEEE, 2023: 1-7.
[69] DAI H X, LI Y W, LIU Z L, et al. AD-AutoGPT: an autonomous GPT for Alzheimer??s disease infodemiology[EB/OL]. [2024-07-19]. https://arxiv.org/abs/2306.10095.
[70] RUNDE B S, ALAPATI A, BAZAN N G. The optimization of a natural language processing approach for the automatic detection of Alzheimer??s disease using GPT embeddings[J]. Brain Sciences, 2024, 14(3): 211.
[71] WU C Y, LIN Z H, FANG W L, et al. A medical diagnostic assistant based on LLM[C]//Proceedings of the 2024 China Health Information Processing Conference. Singapore: Springer, 2024: 135-147.
[72] ULLAH E, PARWANI A, BAIG M M, et al. Challenges and barriers of using large language models (LLM) such as ChatGPT for diagnostic medicine with a focus on digital pathology-a recent scoping review[J]. Diagnostic Pathology, 2024, 19(1): 43.
[73] SCHERBAKOV D, HUBIG N, JANSARI V, et al. The emergence of large language models (LLM) as a tool in literature reviews: an LLM automated systematic review[EB/OL]. [2024-09-22]. https://arxiv.org/abs/2409.04600.
[74] HALTAUFDERHEIDE J, RANISCH R. The ethics of ChatGPT in medicine and healthcare: a systematic review on large language models (LLMs)[J]. NPJ Digital Medicine, 2024, 7: 183.
[75] PENG L, LUO G X, ZHOU S C, et al. An in-depth evaluation of federated learning on biomedical natural language processing for information extraction[J]. NPJ Digital Medicine, 2024, 7: 127.
[76] SONG Y P, ZHANG J H, TIAN Z L, et al. LLM-based privacy data augmentation guided by knowledge distillation with a distribution tutor for medical text classification[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2402.16515.
[77] MILETIC M, SARIYAR M. Assessing the potentials of LLMs and GANs as state-of-the-art tabular synthetic data generation methods[C]//Proceedings of the 2024 International Conference on Privacy in Statistical Databases. Cham: Springer, 2024: 374-389.
[78] BLALOCK D, ORTIZ J J G, FRANKLE J, et al. What is the state of neural network pruning?[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2003.03033.
[79] SONG L, CHEN Y K, YANG S, et al. Low-rank approximation for sparse attention in multi-modal LLMs[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2024: 13763-13773.
[80] XU X H, LI M, TAO C Y, et al. A survey on knowledge distillation of large language models[EB/OL]. [2024-07-20]. https://arxiv.org/abs/2402.13116.
[81] GOU J P, YU B S, MAYBANK S J, et al. Knowledge distillation: a survey[J]. International Journal of Computer Vision, 2021, 129(6): 1789-1819.
[82] KUMAR A, NASEER M, NARAYAN S, et al. Multi-modal generation via cross-modal in-context learning[EB/OL]. [2024-07-19]. https://arxiv.org/abs/2405.18304.
[83] LI Y, SONG Z C, GONG Z J, et al. Multimodal growth and development assessment model[EB/OL]. [2024-10-20]. https:// arxiv.org/abs/2410.13647.
[84] LIU Y S, WU Z Q, CHEN B Z, et al. Medical cross-modal prompt hashing with robust noisy correspondence learning[C]//Proceedings of the 2024 International Conference on Medical Image Computing and Computer Assisted Intervention. Cham: Springer, 2024: 250-261.
[85] WANG P Y, ZHANG H Q, YUAN Y X. MCPL: multi-modal collaborative prompt learning for medical vision-language model[J]. IEEE Transactions on Medical Imaging, 2024, 43(12): 4224-4235.
[86] HUANG H Y, ZHENG O, WANG D D, et al. ChatGPT for shaping the future of dentistry: the potential of multi-modal large language model[J]. International Journal of Oral Science, 2023, 15: 29.
[87] YAN Q, DUAN J W, WANG J X. Multi-modal concept alignment pre-training for generative medical visual question answering[C]//Findings of the Association for Computational Linguistics: ACL 2024. Stroudsburg: ACL, 2024: 5378-5389.