Mathematical Modeling And Explainable Artificial Intelligence For Multivariate Disease Progression Prediction And Personalized Clinical Decision Making In General Medicine And General Surgery

Authors

  • Muruganantham S Author
  • Dr.A. Elakkiya Author
  • Dr.D. Sundaranarayana Author
  • E. Soumya Author
  • A. Swathi Author
  • Dr.K. Sathishkumar Author
  • Dr. Fathimunnisa Author
  • Dr. N. Sandhya Author

Keywords:

Mathematical modeling; Explainable artificial intelligence; Disease progression; Clinical decision support; General medicine; General surgery; Personalized medicine.

Abstract

Accurate prediction of disease progression is crucial when making early and individualised decisions in general medicine and general surgery where patient condition is dynamic and changes in response to a combination of physiological and clinical factors. Existing risk scores partially capture temporal interactions and largely lack clinical interpretation.Existing risk scores have limited ability to represent temporal interaction, and data-driven risk scores are lacking of clinical interpretation. In this study, we suggest an integrated Mathematical Modeling and Explainable Artificial Intelligence (MM-XAI) framework for multivariate disease-progression prediction and personalized clinical decision support. The clinical state variables associated with each etiology, and which are known to be associated with physiological instability, biochemical abnormality, inflammatory burden, disease severity, and clinical/surgical stress, are modeled with nonlinear temporal dynamics. The resulting progression information is then combined with an AI classifier to distinguish between stable, moderate-risk and high-risk states and to provide explanation of patient-specific predictions using a feature attribution algorithm, SHAP. The performance is assessed by accuracy, sensitivity, specificity, F1-score, AUROC, calibration error, progression prediction error and explanation consistency. The framework extends from patient state to mathematical progression to AI prediction to explanation to personalized decision support to create an interpretable computational framework for risk assessment. Real-world clinical deployment requires prospective multicenter clinical trials to validate in the real world.

Downloads

Download data is not yet available.

Downloads

Published

2026-09-22

How to Cite

Mathematical Modeling And Explainable Artificial Intelligence For Multivariate Disease Progression Prediction And Personalized Clinical Decision Making In General Medicine And General Surgery. (2026). Minimax Theory and Its Applications, 11(2s), 277-285. https://journalmta.com/index.php/jmta/article/view/415