Mapping Gene Expression and Transcription Inhibition to Cancer Treatment Cost Estimation

The escalating costs of cancer treatments present a significant challenge for healthcare systems globally. Traditional methods of estimating cancer treatment costs often fail to incorporate patient-specific biological factors. Recent advancements in predictive oncology have emphasized the need for machine learning techniques to predict treatment costs based on biological attributes. Transcription inhibition plays a pivotal role in regulating gene expression and cancer cell proliferation, yet the relationship between transcription inhibition and treatment budgets remains under- explored. This study aims to address this gap by creating a machine learning (ML) framework that combines transcription inhibition dynamics with cancer treatment budgeting, with the goal of improving the accuracy of cost estimations. Our framework involves generating synthetic data representing transcription inhibition attributes, training multiple regression models, and evaluating them based on metrics like Mean Absolute Error (MAE), Mean Squared Error (MSE), and the R² score. The machine learning models applied include Linear Regression, Random Forest, Gradient Boosting, Support Vector Regressor, and Neural Network Regressor. Random Forest and Gradient Boosting models showed promising results, outperforming other models in budget prediction. This approach offers potential applications in precision oncology, with implications for optimizing resource allocation and supporting cost-effective clinical decision-making.