Stochastic Mathematical Models and Machine Learning Algorithms for Early Detection of Cancer and Diabetes

Abstract

Early detection of chronic diseases such as cancer and diabetes is essential for improving patient survival, reducing treatment costs, and supporting timely clinical decision-making. However, disease progression is inherently uncertain due to biological variability, genetic factors, environmental influences, and lifestyle changes, making accurate prediction a challenging task. This study proposes a generalized hybrid framework that integrates stochastic mathematical models with machine learning algorithms for the early detection of cancer and diabetes. Initially, stochastic differential equations are employed to model the uncertain evolution of disease-related variables, such as blood glucose levels for diabetes and tumor growth for cancer. Synthetic patient datasets are then generated by combining demographic, clinical, and stochastic features. These datasets are used to train an Artificial Neural Network (ANN) classifier capable of distinguishing between healthy and diseased individuals. The proposed framework is evaluated using standard performance metrics, including accuracy, precision, recall, F1-score, and ROC-AUC. The integration of stochastic disease progression with machine learning enables the prediction model to capture both temporal uncertainty and complex nonlinear relationships among clinical variables. Furthermore, the framework is sufficiently general to be extended to multiple chronic diseases by modifying only the disease-specific variables while preserving the same computational pipeline. The proposed methodology offers a mathematically interpretable and computationally efficient approach for intelligent healthcare systems, clinical decision support, and personalized medicine.