Selecting Biomarkers for building optimal treatment selection rules using Kernel Machines
This work addresses the challenge of cost-effective biomarker selection for treatment rules in medical applications, representing an incremental improvement by adding cost considerations to existing optimization methods.
The paper tackles the problem of selecting optimal biomarker combinations for treatment selection by minimizing total disease and treatment burden, while also incorporating biomarker measurement costs to avoid expensive and performance-hurting models. The result demonstrates the importance of including feature selection and marker costs in deriving treatment-selection rules through simulations and a real data example.
Optimal biomarker combinations for treatment-selection can be derived by minimizing total burden to the population caused by the targeted disease and its treatment. However, when multiple biomarkers are present, including all in the model can be expensive and hurt model performance. To remedy this, we consider feature selection in optimization by minimizing an extended total burden that additionally incorporates biomarker measurement costs. Formulating it as a 0-norm penalized weighted classification, we develop various procedures for estimating linear and nonlinear combinations. Through simulations and a real data example, we demonstrate the importance of incorporating feature-selection and marker cost when deriving treatment-selection rules.