Detailed predictive modeling of hospital readmission risk using a class-balanced machine learning approach
G. Siva Shanmugamgsivashanmugam@rvu.edu.inDepartment of Computer EngineeringSchool of Computer Science & EngineeringRV UniversityBangalore, Karnataka, 560059, IndiaView full profile → , *Akarsh KunwarCorresponding authorakarsh.kunwar@gmail.comDepartment of Computer EngineeringSchool of Computer Science and EngineeringVellore Institute of TechnologyVellore, Tamil Nadu, 632014, IndiaView full profile → , M. Anbarasimanbarasi@vit.ac.inDepartment of Computer EngineeringSchool of Computer Science and EngineeringVellore Institute of TechnologyVellore, Tamil Nadu, 632014, IndiaView full profile → , Ojas Nahtaojasnahta2004@gmail.comDepartment of Computer EngineeringSchool of Computer Science and EngineeringVellore Institute of TechnologyVellore, Tamil Nadu, 632014, IndiaView full profile → , Vrinda Kumarvrindakr0504@gmail.comDepartment of Computer EngineeringSchool of Computer Science and EngineeringVellore Institute of TechnologyVellore, Tamil Nadu, 632014, IndiaView full profile → , Dhruv Jaintodhruvjain@gmail.comDepartment of Computer EngineeringSchool of Computer Science and EngineeringVellore Institute of TechnologyVellore, Tamil Nadu, 632014, IndiaView full profile →
* Corresponding author · click or hover a name for details
- Received:
- 01 Jun 2026
- Published Online:
- 30 Sep 2026
- Article type:
- Research Article
- Language:
- EN
- Article no.:
- JIOS-2448
- Pages:
- 3509–3519
Abstract
Forecasting 30-day hospital readmission is a major concern in healthcare because of its financial cost and the missed opportunity to act early on high-risk patients. We highlight the accuracy trap inherent in naive modeling of highly imbalanced data: a typical SMOTE-based Random Forest baseline reached a misleading 91% accuracy yet identified almost no high-risk patients (recall ≈ 0.0%). To address this, we built a comparative methodology that moves from cost-sensitive Random Forest classifiers to an optimized final model: XGBoost with class-balancing weights and post-hoc threshold tuning (lowered to 0.4396). The final model achieved a readmission recall of 69.97%, far above the baseline, offering a viable basis for clinical decision support in which identifying at-risk patients is the priority.
Keywords
Subject Classifications
References
[1] A. Kunwar, “Readmission Predictor,” GitHub. [Online]. Available: https://github.com/akarshkunwar/ReadmissionPredictor (2026).
[2] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), pp. 4765–4774 (2017).
[3] L. Breiman, “Random forests,” Mach. Learn., vol. 45, no. 1, pp. 5–32 (2001).
[4] H. He and E. A. Garcia, “Learning from imbalanced data,” IEEE Trans. Knowl. Data Eng., vol. 21, no. 9, pp. 1263–1284 (2009).
[5] B. Strack, J. P. DeShazo, E. A. Chrischilles, and B. Leifson, “Impact of HbA1c measurement on hospital readmission rates for patients with diabetes,” BioMed Res. Int., vol. 2014, Art. no. 781670 (2014).
[6] D. Kansagara, H. Englander, A. Salanitro, D. Kagen, C. Theobald, M. Freeman, and S. Kripalani, “Risk prediction models for hospital readmission: A systematic review,” JAMA, vol. 306, no. 15, pp. 1688–1698 (2011).
[7] S. F. Jencks, M. V. Williams, and E. A. Coleman, “Rehospitalizations among patients in the Medicare fee-for-service program,” N. Engl. J. Med., vol. 360, no. 14, pp. 1418–1428 (2009).
[8] Y. Huang, A. Talwar, S. Chatterjee, and R. R. Aparasu, “Application of machine learning in predicting hospital readmissions: A scoping review of the literature,” BMC Med. Res. Methodol., vol. 21, Art. no. 96 (2021).
[9] Y. Shang, K. Jiang, L. Wang, Z. Zhang, S. Zhou, Y. Liu, J. Dong, and H. Wu, “The 30-days hospital readmission risk in diabetic patients: Predictive modeling with machine learning classifiers,” BMC Med. Inform. Decis. Mak., vol. 21, no. Suppl. 2, Art. no. 57 (2021).
[10] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Min. pp. 785–794 (2016).
[11] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: Synthetic minority over-sampling technique,” J. Artif. Intell. Res., vol. 16, pp. 321–357 (2002).
[12] F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, “Scikit-learn: Machine learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830 (2011).
[13] G. Lemaître, F. Nogueira, and C. K. Aridas, “Imbalanced-learn: A Python toolbox to tackle the curse of imbalanced datasets in machine learning,” J. Mach. Learn. Res., vol. 18, no. 17, pp. 1–5 (2017).
[14] B. Strack, J. P. DeShazo, C. Gennings, J. L. Olmo, S. Ventura, K. J. Cios, and J. N. Clore, “Impact of HbA1c measurement on hospital readmission rates: Analysis of 70,000 clinical database patient records,” BioMed Res. Int., vol. 2014, pp. 1–11 (2014).
[15] S. Purushothaman, G. S. Shanmugam, and S. Nagarajan, “Achieving seamless semantic interoperability and enhancing text embedding in Healthcare IoT: A deep learning approach with survey,” SN Comput. Sci., vol. 5, p. 99 (2024).
[16] P. Sony, G. Siva Shanmugam, and N. Sureshkumar, “Enhancement of semantic interoperability in healthcare systems using IFCIoT architecture,” KSII Trans. Internet Inf. Syst., vol. 18, no. 4, pp. 881–902 (2024).
[17] P. Sony, G. Siva Shanmugam, and N. Sureshkumar, “A precise IFCIoT platform for the achievement of semantic interoperability in healthcare data systems,” in Proc. Int. Conf. Artif. Intell. Softw. Eng. (ICAISE), Kochi, India, pp. 29–34 (2023).
[18] P. Sony, G. S. Shanmugam, and S. Nagarajan, “An intuitionistic fuzzy-based intelligent system for semantic interoperability and privacy preservation in healthcare systems,” Soft Computing, (2023), doi: 10.1007/s00500-023-08972-6.
[19] M. Anbarasi and M. A. Saleemdurai, “Social networks based disease analysis using Hadoop,” Int. J. Pharm. Technol., vol. 8, no. 2, pp. 4130–4141 (2016).
[20] U. Tiwari, M. Anbarasi, and M. A. Saleemdurai, “A rough set based kernel possibilistic C-means algorithm for medical data,” Int. J. Pharm. Technol., vol. 8, no. 2, pp. 16596–16604 (2016).
[21] S. Gupta and A. Chug, “Software maintainability prediction using an enhanced random forest algorithm,” J. Discrete Math. Sci. Cryptogr., vol. 23, no. 2, pp. 565–580 (2020).
[22] A. Gupta, M. C. Lohani, and M. Manchanda, “Financial fraud detection using naive Bayes algorithm in highly imbalance data set,” J. Discrete Math. Sci. Cryptogr., vol. 24, no. 5, pp. 1559–1572 (2021).
