Premium
Selecting biomarkers for building optimal treatment selection rules by using kernel machines
Author(s) -
Dasgupta Sayan,
Huang Ying
Publication year - 2020
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/rssc.12379
Subject(s) - feature selection , selection (genetic algorithm) , computer science , mathematical optimization , kernel (algebra) , population , norm (philosophy) , machine learning , data mining , artificial intelligence , mathematics , medicine , combinatorics , environmental health , political science , law
Summary Optimal biomarker combinations for treatment selection can be derived by minimizing the 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 can hurt model performance. To remedy this, we consider feature selection in optimization by minimizing an extended total burden that additionally incorporates biomarker costs. Formulating it as a 0‐norm penalized weighted classification, we develop various procedures for estimating linear and non‐linear combinations. Through simulations and a real data example, we demonstrate the importance of incorporating feature selection and marker cost when deriving treatment selection rules.