A1112
Title: Linear biomarker combination for constrained classification
Authors: Yijian Huang - Emory University (United States) [presenting]
Abstract: Multiple biomarkers are often combined to improve disease diagnosis. Unfortunately, the uniformly optimal combination, i.e., with respect to all reasonable performance metrics, requires excessive distributional modelling, to which the estimation can be sensitive. An alternative strategy is rather to pursue local optimality concerning a specific performance metric. Nevertheless, existing methods may not target the clinical utility of the intended medical test, which usually needs to operate above a certain sensitivity or specificity level, or do not have their statistical properties well studied and understood. The developments and investigation of a linear combination method will be discussed to maximize the clinical utility empirically for such a constrained classification. The combination coefficient is shown to have cube root asymptotics. The convergence rate and limiting distribution of the predictive performance are subsequently established, exhibiting the robustness of the method in comparison with others. An algorithm with sound statistical justification is devised for efficient and high-quality computation. Simulations corroborate the theoretical results and demonstrate good statistical and computational performance. An illustration with a clinical study on aggressive prostate cancer detection is provided.