Abstract
An artificial intelligence-based predictive model was developed for the early identification of postoperative complications using clinical and functional indicators. A total of 10,000 synthetic records from the TERMINET dataset were analyzed using logistic regression, Random Forest, and XGBoost, with stratified data splitting, cross-validation, and discrimination and calibration metrics. The models demonstrated exceptional predictive performance, with perfect classification achieved by logistic regression and Random Forest. SHAP analysis identified sodium, urea nitrogen, creatinine, and INR-related parameters among the main predictors. The findings demonstrate the computational feasibility of the proposed approach; however, the exceptionally high performance warrants cautious interpretation and subsequent external validation using real and independent clinical cohorts.
References
[2] K. Merath, J. M. Hyer, R. Mehta et al., “Use of machine learning for prediction of patient risk of postoperative complications after liver, pancreatic, and colorectal surgery,” Journal of Gastrointestinal Surgery, vol. 24, pp. 1843–1851, 2020, doi: 10.1007/s11605-019-04338-2.
[3] P. Thottakkara, T. Ozrazgat-Baslanti, B. B. Hupf, P. Rashidi, P. Pardalos, P. Momcilovic, and A. Bihorac, “Application of machine learning techniques to high-dimensional clinical data to forecast postoperative complications,” PLoS ONE, vol. 11, no. 5, Art. no. e0155705, 2016, doi: 10.1371/journal.pone.0155705.
[4] A. M. Hassan, A. Rajesh, M. Asaad, J. A. Nelson, J. H. Coert, B. J. Mehrara, and C. E. Butler, “Artificial intelligence and machine learning in prediction of surgical complications: Current state, applications, and implications,” The American Surgeon, vol. 89, no. 1, pp. 25–30, 2023.
[5] S. Kokkinakis, E. I. Kritsotakis, and K. Lasithiotakis, “Artificial intelligence in surgical risk prediction,” Journal of Clinical Medicine, vol. 12, no. 12, Art. no. 4016, 2023.
[6] C. Varghese, E. M. Harrison, G. O’Grady, and E. J. Topol, “Artificial intelligence in surgery,” Nature Medicine, vol. 30, no. 5, pp. 1257–1268, 2024, doi: 10.1038/s41591-024-02970-3.
[7] A. Guni, P. Varma, J. Zhang, M. Fehervari, and H. Ashrafian, “Artificial intelligence in surgery: The future is now,” European Surgical Research, vol. 65, no. 1, pp. 22–39, 2024.
[8] A. Bonde, K. M. Varadarajan, N. Bonde et al., “Assessing the utility of deep neural networks in predicting postoperative surgical complications: A retrospective study,” The Lancet Digital Health, vol. 3, no. 8, pp. e471–e485, 2021.
[9] S. Chen, T. Deng, Q. Yang et al., “Development and validation of an explainable machine learning model for predicting postoperative pulmonary complications after lung cancer surgery: A machine learning study,” EClinicalMedicine, vol. 86, 2025.
[10] H. Deng, Z. Eftekhari, C. Carlin et al., “Development and validation of an explainable machine learning model for major complications after cytoreductive surgery,” JAMA Network Open, vol. 5, no. 5, Art. no. e2212930, 2022.
[11] P. Li, S. Gao, Y. Wang et al., “Utilising intraoperative respiratory dynamic features for developing and validating an explainable machine learning model for postoperative pulmonary complications,” British Journal of Anaesthesia, vol. 132, no. 6, pp. 1315–1326, 2024.
[12] N. Kenig, J. Monton Echeverria, and A. Muntaner Vives, “Artificial intelligence in surgery: A systematic review of use and validation,” Journal of Clinical Medicine, vol. 13, no. 23, Art. no. 7108, 2024.
[13] W. T. Stam, L. K. Goedknegt, E. W. Ingwersen, L. J. Schoonmade, E. R. Bruns, and F. Daams, “The prediction of surgical complications using artificial intelligence in patients undergoing major abdominal surgery: A systematic review,” Surgery, vol. 171, no. 4, pp. 1014–1021, 2022.
[14] T. J. Loftus, P. J. Tighe, A. C. Filiberto et al., “Artificial intelligence and surgical decision-making,” JAMA Surgery, vol. 155, no. 2, pp. 148–158, 2020.
[15] D. Lekka, A. Pnevmatikakis, E. Kanavos, B. Gottardelli, A. M. Tudor, P. Cornacchione, M. de Angeli, and A. Bellieni, “TERMINET eHealth post-operation complications synthetic dataset,” Zenodo, Dataset, 2024, doi: 10.5281/zenodo.10885957.
This work is licensed under a Creative Commons Attribution 4.0 International License.