Abstract
Multimodal perioperative data from patients undergoing atrial fibrillation (AF) ablation offer valuable insights for stratifying recurrence risk, yet remain underutilized in prediction models. This multicenter retrospective study included 2508 patients who underwent AF ablation at five Chinese centers: The Fourth Affiliated Hospital of Zhejiang University School of Medicine (Jan 2016–Mar 2024; Training Cohort), Taizhou Hospital of Zhejiang Province (Jan 2015–Jan 2024; Training Cohort), The Affiliated Hospital of Yunnan University (Jan 2016–Jan 2024; Validation Cohort), Jinhua People’s Hospital (Jan 2020–Jan 2024; Test Cohort), and Ningbo Beilun Hospital (Jan 2020–Jan 2024; Test Cohort). We developed a dual-branch deep learning model to predict AF recurrence, in which structured data were processed via a 1D ResNet and textual data were encoded using four large language models (LLaMA-7B, Phi2-2.7B, Mistral-7B, and MedGemma-27B). The model incorporating MedGemma for text feature extraction performed best, achieving areas under the curve of 0.934 (95% CI: 0.921–0.946), 0.928 (95% CI: 0.904–0.950), and 0.911 (95% CI: 0.878–0.941) on the training, validation, and test sets, respectively. Our model integrates multimodal perioperative data from AF ablation patients, effectively identifies high-risk individuals, and may facilitate targeted interventions to reduce relapse.
Data availability
The datasets generated and/or analyzed during the current study are not publicly available as they contain confidential patient information, but are available from the corresponding author upon reasonable request.
References
Writing Committee Members et al 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J. Am. Coll. Cardiol. 83, 109–279 (2024).
Zink, M. D. et al. Predictors of recurrence of atrial fibrillation within the first 3 months after ablation. Europace. 22, 1337–1344 (2020).
Dretzke, J. et al. Predicting recurrent atrial fibrillation after catheter ablation: a systematic review of prognostic models. Europace. 22, 748–760 (2020).
Jia, S. et al. Association between triglyceride-glucose index trajectories and radiofrequency ablation outcomes in patients with stage 3D atrial fibrillation. Cardiovasc. Diabetol. 23, 121 (2024).
Black-Maier, E. et al. Predicting atrial fibrillation recurrence after ablation in patients with heart failure: validity of the APPLE and CAAP-AF risk scoring systems. Pacing Clin. Electrophysiol. 42, 1440–1447 (2019).
Shool, S. et al. A systematic review of large language model (LLM) evaluations in clinical medicine. BMC Med. Inform. Decis. Mak. 25, 117 (2025).
Google. MedGemma. GitHub repository, https://github.com/Google-Health/medgemma (2025).
Microsoft. Phi-2: The surprising power of small language models. Microsoft Research Blog, https://www.microsoft.com/en-us/research/blog/phi-2-the-surprising-power-of-small-language-models/ (2023).
Touvron, H. et al. Llama 2: open foundation and fine-tuned chat models. Preprint at arXiv https://doi.org/10.48550/arXiv.2307.09288 (2023).
Jiang, A. Q. et al. Mistral 7B. Preprint at arXiv https://doi.org/10.48550/arXiv.2310.06825 (2023).
Bedi, S. et al. Testing and evaluation of health care applications of large language models: a systematic review. JAMA. 333, 319–328 (2025).
Parameswaran, R., Al-Kaisey, A. M. & Kalman, J. M. Catheter ablation for atrial fibrillation: current indications and evolving technologies. Nat. Rev. Cardiol. 18, 210–225 (2021).
Brundel, B. J. J. M. et al. Atrial fibrillation. Nat. Rev. Dis. Primers 8, 21 (2022).
Hu, Y. F., Chen, Y. J., Lin, Y. J. & Chen, S. A. Inflammation and the pathogenesis of atrial fibrillation. Nat. Rev. Cardiol. 12, 230–243 (2015).
Krasteva, V. et al. Detection of atrial fibrillation in Holter ECG recordings by ECHOView images: a deep transfer learning study. Diagnostics 15, 865 (2025).
Zhang, P. et al. Automatic screening of patients with atrial fibrillation from 24-h Holter recording using deep learning. Eur. Heart J. Digit. Health 4, 216–224 (2023).
Jia, S. et al. A simple logistic regression model for predicting the likelihood of recurrence of atrial fibrillation within 1 year after initial radio-frequency catheter ablation therapy. Front. Cardiovasc. Med. 8, 819341 (2022).
Sohns, C. & Marrouche, N. F. Atrial fibrillation and cardiac fibrosis. Eur. Heart J. 41, 1123–1131 (2020).
Faber, J. & Fonseca, L. M. How sample size influences research outcomes. Dental Press J. Orthod. 19, 27–29 (2014).
Rajput, D., Wang, W. J. & Chen, C. C. Evaluation of a decided sample size in machine learning applications. BMC Bioinform. 24, 48 (2023).
Acknowledgements
The authors are grateful to the PixelmedAI platform (https://github.com/410312774/PixelMedAI) for providing technical support for the deep learning network in this study. This study was supported by Key Projects of the Zhejiang Provincial Natural Science Foundation (No. LZ25H020001), Joint TCM Science & Technology Projects of National Demonstration Zones for Comprehensive TCM Reformt (No. GZY-KJS-ZJ-2025-022), National Natural Science Foundation of China (No. 81971688), Scientific Research Fund of Zhejiang Provincial Education Department (No. Y202457057) and Taizhou Science and Technology Project (No. 21ywa02).
Author information
Authors and Affiliations
Contributions
Conceptualization: J.S., Y.Y, X.S. Methodology: J.S., L.C. Software: J.S., L.C. Validation: Z.Q., Y.Y., Z.H., Z.J. Formal analysis: J.S. Investigation: M.X., Y.Q., Z.J. Resources: Y.Y., G.Y., Y.X., S.P., L.J., X.Q. Data curation: J.S, Y.Y., G.Y., Y.X., S.P. Writing—original draft: J.S. Writing—review and editing: J.S., X.S. Visualization: L.W, F.C., Y.Y. Supervision: X.S., G.W. Project administration: X.S., G.W. Funding acquisition: J.S., X.S., Y.Y.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Jia, S., Yin, Y., Guan, Y. et al. A deep learning model integrating structured data and clinical text for predicting atrial fibrillation recurrence. npj Digit. Med. (2026). https://doi.org/10.1038/s41746-026-02436-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41746-026-02436-5
