Progress and Prospects of Artificial Intelligence Application in the DRG/DIP

doi:10.19546/j.issn.1674-3830.2026.9.007

China Health Insurance ›› 2026, Vol. 0 ›› Issue (9) : 57-66. DOI: 10.19546/j.issn.1674-3830.2026.9.007

Observation & Discussion

Progress and Prospects of Artificial Intelligence Application in the DRG/DIP

Author information +

History +

Abstract

DRG/DIP faces multiple challenges in both grouping and admission management, including large-scale datasets, variable medical record quality, reliance on manual expertise for rule setting, and high technical demands for data processing. Admission management further involves complex tasks such as detecting enrollment errors and abnormal behaviors, which directly impact the fairness and efficiency of medical insurance payments. Artificial intelligence (AI) offers promising solutions and technologies to these challenges. By efficiently processing large volumes of medical data, AI enhances data utilization, while modeling and learning from historical records allows it to capture complex patterns of medical behavior. This capability supports more accurate development of grouping schemes, admission management, and dynamic adjustment, thereby improving the scientific rigor and intelligence of the payment system. Focusing on DRG as a representative example, this paper examines the core difficulties of traditional, manual-driven medical insurance management, highlights the advantages and characteristics of AI in this domain, and reviews domestic and international research on its applications in medical behavior feature extraction, grouping optimization, and admission supervision. The approaches discussed are also applicable to DIP and other payment models, providing insights for the intelligent transformation of China’s medical insurance payment system.

Key words

reform of medical insurance payment / artificial intelligence / DRG / DIP

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Progress and Prospects of Artificial Intelligence Application in the DRG/DIP[J]. China Health Insurance. 2026, 0(9): 57-66 https://doi.org/10.19546/j.issn.1674-3830.2026.9.007

References

[1] Cam Donaldon, Jon Magnussen.DRGs: The road to hospital efficiency[J]. Health Policy, 1992, 21(1): 47-64.
[2] Busse R, Geissler A, Aaviksoo A, et al.Diagnosis related groups in Europe: moving towards transparency, efficiency, and quality in hospitals?[J]. BMJ, 2013:346.
[3] Eugene Levine, Faye G.Abdellah. DRGs: A recent refinement to an old method[J]. Inquiry, 1984, 21(2): 105-112.
[4] 周宇,郑树忠,孙国桢.德国DRG付费制度的借鉴[J].中国卫生资源,2004,7(4):186-187.
[5] 王亦冬,孙志楠,陈颖.典型国家DRG研究与实践进展综述及其对我国的启示[J].中国卫生经济,2021,40(6):91-96.
[6] 姚奕婷,翁一冰,邓婕,等.DRG发展与应用回顾及付费标准研究综述[J].中国卫生经济,2018,37(1):24-27.
[7] 吴晶,赵博雅,朱玄,等.DRG分组策略与支付标准界定的国内外比较[J].中国医疗保险,2021(02):75-80.
[8] 国家医疗保障疾病诊断相关分组(CHS-DRG)分组与付费技术规范[EB/OL].(2019-10-16)[2026-05-01]. https://www.gov.cn/zhengce/zhengceku/2019-11/18/5562261/files/d1e38788efee4493b732fdcb0776a2b7.pdf.
[9] 张惠娟,陈佩贞.疾病诊断及其编码的准确性是实施DRGs支付方式的关键[J].中医药管理杂志,2013,21(02):197-199.
[10] 赵佳,杨晓琳,徐长妍.76份恶性肿瘤住院病案编码缺陷对DRG分组的影响[J].中国病案,2021,22(10):35-37.
[11] 昆的巴哈尔·阿仁,丁玉梅.神经外科住院病案编码缺陷对DRG分组的影响策略[J].特别健康,2024(4):134-135.
[12] 黄华波. 找准政策立足点——完善制度防范基金风险[J].中国医疗保险,2020(8):29-30.
[13] 黄秋华,罗小丽.开展DRGS对提升病案编码员肿瘤疾病分类能力的相关对策[J].医学理论与实践,2018,31(20):3157-3158.
[14] 黎俊,周王艳,唐再丽.DRG背景下病案科的精细化管理[J].中国病案,2024,25(05):6-8.
[15] 刘路明,李盈,金雯,等.病案首页全流程质控体系构建及应用[J].现代医院,2024,24(11):1722-1725.
[16] 李水娥,丁锦希,李佳明,等.DRG创新除外机制的关键问题研究[J].中国医疗保险,2023(04):41-48.
[17] Brown T, Mann B, Ryder N, et al.Language models are few-shot learners[J]. Advances in Neural Information Processing Systems, 2020,33: 1877-1901.
[18] OpenAI. GPT-4 Technical Report[J]. arXiv preprint arXiv:2303.08774, 2023.
[19] 浙江金华广福肿瘤医院.融合深度学习和知识图谱的病案自动编码及DRG运行分析系统研究[EB/OL].(2023-10-20)[2026-09-21].https://www.chima.org.cn/Html/News/Articles/16536.html.
[20] 张俊.科大讯飞:AI赋能医保 “惠及”你我他[EB/OL].(2022-09-27)[2026-05-21].https://www.ah.chinanews.com.cn/news/2022/0927/310182.shtml.
[21] 新华日报.江苏已建成医保反欺诈模型:覆盖45个场景全流程闭环[EB/OL].(2026-04-08)[2026-05-21].https://www.jiangsu.gov.cn/art/2026/4/8/art_60085_11755087.html.
[22] DRG变量.徐州开辟DRG监管新路径[EB/OL].(2023-08-18)[2026-09-21].https://www.cn-healthcare.com/articlewm/20230818/content-1593697.html.
[23] Johnson A E W, Pollard T J, Shen L, et al. MIMIC-III, a freely accessible critical care database[J]. Scientific Data,2016,3:160035.
[24] Johnson A E W, Bulgarelli L, Shen L, et al. MIMIC-IV, a freely accessible electronic health record dataset[J]. Scientific Data,2023,10(1):1.
[25] Pollard T J, Johnson A E W, Raffa J D, et al. The eICU Collaborative Research Database, a freely available multi-center database for critical care research[J]. Scientific Data,2018,5(1):1-13.
[26] Dernoncourt F, Lee J Y.Pubmed 200k RCT: a dataset for sequential sentence classification in medical abstracts[C]//Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 2: Short Papers).2017:308-313.
[27] Mujtaba G, Shuib L, Raj R G, et al.Automatic ICD-10 multi-class classification of cause of death from plaintext autopsy reports through expert-driven feature selection[J]. PLoS One,2017,12(2):e0170242.
[28] Berndorfer S, Henriksson A.Automated Diagnosis Coding with Combined Text Representations[J]. Studies in Health Technology and Informatics,2017, 235:201-205.
[29] Medori J, Fairon C.Machine learning and features selection for semi-automatic ICD-9-CM encoding[C]//Proceedings of the NAACL HLT 2010 Second Louhi Workshop on Text and Data Mining of Health Documents.2010: 84-89.
[30] Erraguntla M, Gopal B, Ramachandran S, et al.Inference of missing ICD 9 codes using text mining and nearest neighbor techniques[C]//2012 45th Hawaii International Conference on System Sciences.IEEE,2012:1060-1069.
[31] Mullenbach J, Wiegreffe S, Duke J, et al.Explainable prediction of medical codes from clinical text[C]//Proceedings of NAACL-HLT.2018:1101-1111.
[32] Chen J, Teng F, Ma Z, et al.A multi-channel convolutional neural network for ICD coding[C]//2019 IEEE 14th International Conference on Intelligent Systems and Knowledge Engineering (ISKE).IEEE,2019:1178-1184.
[33] 王阳阳,郑西川.基于自注意力机制的CNN-LSTM模型在ICD智能编码系统中的应用研究[J].中国数字医学,2020,15(11):20-24.
[34] Wang H, Gao C, Dantona C, et al.DRG-LLaMA: tuning LLaMA model to predict diagnosis-related group for hospitalized patients[J].NPJ Digital Medicine,2024,7(1):16.
[35] 张艺,滕飞,胡节.基于多任务学习的国际疾病分类自动编码模型[J].广西科学,2023,30(1):114-120.
[36] Lee J, Yoon W, Kim S, et al.BioBERT: a pre-trained biomedical language representation model for biomedical text mining[J]. Bioinformatics,2020,36(4): 1234-1240.
[37] Heo T S, Yoo Y, Park Y, et al.Medical code prediction from discharge summary: Document to sequence BERT using sequence attention[C]//2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA). IEEE,2021:1239-1244.
[38] Strydom S.Automatic assignment of diagnosis codes to free-form text medical notes[D]. Stellenbosch: Stellenbosch University,2021.
[39] 陈惠荷,郑威林,陈卓健.同义词约束库在智能化编码解决方案的研究[J].智慧健康,2024,10(13):6-8+12.
[40] 朱明宇. 基于医学人工智能技术的病案智能编码研究[J].中国数字医学,2018,13(4):34-36.
[41] Bai T, Vucetic S.Improving medical code prediction from clinical text via incorporating online knowledge sources[C]//The World Wide Web Conference. ACM, 2019: 72-82.
[42] Yuan Z, Tan C, Huang S.Code synonyms do matter: Multiple synonyms matching network for automatic ICD coding[C]//ACL. 2022.
[43] 陈志添. 基于决策树的诊断相关组分类研究[D].广州:华南理工大学,2018.
[44] 吴尔律,梁洪彬,罗海琼,等.基于机器学习的DRG分组预测研究[J].信息与电脑,2023,35(5):22-25.
[45] 陈龙,马利,何文英,等.高血压住院患者DRGs分组方法研究[J].中国卫生统计,2015,32(01):110-111.
[46] Kass G V.An exploratory technique for investigating large quantities of categorical data[J]. Journal of the Royal Statistical Society: Series C (Applied Statistics),1980, 29(2): 119-127.
[47] 李晓雨,李文瑾,田立启.基于CHAID算法的肝癌患者住院费用DRGs分组及超限费用影响因素研究[J].中国社会医学杂志,2023,40(04):497-501.
[48] 孙梗钤,卿勇,张眆凯,等.基于决策树和结构方程模型的HIV/AIDS患者CHS-DRG分组效果检验[J].现代预防医学,2023,50(08):1499-1503.
[49] 彭伟彬,刘颜,鲍俊翠,等.基于通径分析和决策树模型的异位妊娠手术患者病例组合研究[J].中国卫生经济,2018,37(12):91-93.
[50] Structural equations with latent variables[M]. John Wiley & Sons, 1989.
[51] Wright S.The method of path coefficients[J]. The Annals of Mathematical Statistics, 1934,5(3):161-215.
[52] Gartner D, Kolisch R, Neill D B, et al.Machine learning approaches for early DRG classification and resource allocation[J]. INFORMS Journal on Computing, 2015,27(4):718-734.
[53] 张岩曦,刘晓聪,杨杰,等.一种融合ICD层级特征的DRG自动分组方法及系统:CN202211094495.X[P].2023-05-16.
[54] Islam M M, Li G H, Poly T N, et al.DeepDRG: Performance of artificial intelligence model for real-time prediction of diagnosis-related groups[J]. Healthcare, 2021,9(12):1632.
[55] Liu J, Capurro D, Nguyen A, et al.Early prediction of diagnostic-related groups and estimation of hospital cost by processing clinical notes[J]. NPJ Digital Medicine, 2021,4(1):103.
[56] McKinzie B, Gan Z, Fauconnier J P, et al. MM1: methods, analysis and insights from multimodal LLM pre-training[C]//European Conference on Computer Vision. Cham: Springer Nature Switzerland, 2024:304-323.
[57] Alsentzer E, Murphy J R, Boag W, et al. Publicly available clinical BERT embeddings[EB/OL]. (2019-04-06)[2026-05-21].https://arxiv.org/abs/1904.03323v1.
[58] 南方网.科大讯飞DRG分组AI智慧监管方案获得“精确医保决策赛道”第一名[EB/OL].(2022-03-09)[ 2026-05-21]. https://tech.southcn.com/node_1da01ca84b/3731ce870d.shtml.
[59] Centers for Medicare & Medicaid Services. Basic Stand Alone (BSA) Medicare Claims Public Use Files (PUFs)[EB/OL].(2026-07-20)[2026-05-21]. https://www.cms.gov/data-research/statistics-trends-and-reports/basic-stand-alone-medicare-claims-public-use-files.
[60] Gupta R A. Healthcare provider fraud detection analysis[EB/OL].(2019-05-09)[2026-05-22].https://www.kaggle.com/datasets/rohitrox/healthcare-provider-fraud-detection-analysis.
[61] Massi M C, Ieva F, Lettieri E.Data mining application to healthcare fraud detection: a two-step unsupervised clustering method for outlier detection with administrative databases[J]. BMC Medical Informatics and Decision Making,2020,20:1-11.
[62] 卢洪满. 基于无监督算法的DRG细分组在编码高套上的应用研究[J].数字通信世界,2022(12):117-119.
[63] Shekhar S, Leder-Luis J, Akoglu L.Unsupervised machine learning for explainable health care fraud detection[R].No.w30946. National Bureau of Economic Research, 2023.
[64] Chen R, Zhang H, Lin K.A graph-based method for health care joint fraud detection[C]//Proceedings of the 2020 9th International Conference on Computing and Pattern Recognition. 2020:122-129.
[65] Yoo Y, Shin J, Kyeong S.Medicare fraud detection using graph analysis: A comparative study of machine learning and graph neural networks[J]. IEEE Access, 2023,11:88278-88294.
[66] Hong B, Lu P, Xu H, et al.Health insurance fraud detection based on multi-channel heterogeneous graph structure learning[J]. Heliyon,2024,10(9).
[67] Xiao S, Bai T, Cui X, et al.A graph-based contrastive learning framework for medicare insurance fraud detection[J]. Frontiers of Computer Science, 2023,17(2):172341.
[68] Zhang R, Cheng D, Yang J, et al.Pre-trained online contrastive learning for insurance fraud detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2024,38(20):22511-22519.
[69] 中国药品监管网.广西钦州医保局构建“高编高套”模型堵漏洞[EB/OL]. (2025-02-08)[2026-05-06].http://www.drugnet.com.cn/news/detail/226930.
[70] 烟台晚报.智能系统揪出“过度医疗”[EB/OL].(2025-07-07)[2026-05-01].https://www.shm.com.cn/szb/ytwb/paper/pad/content/202507/07/content_235602.html.
[71] 耿韬.上海:智能监管推动医疗机构高质量发展——读懂2.0时代的医保基金监管[EB/OL]. (2023-09-05)[2026-05-21].https://ybzffs.com/page41?article_id=25946.
[72] 国家医保局. 2024年全国医疗保障事业发展统计公报[EB/OL]. (2025-07-14)[2026-05-01].https://www.nhsa.gov.cn/art/2025/7/14/art_7_17248.html.