A New Measure of Diagnostic Accuracy With Cut-Points Criterion for k-Stage Classification Disease Based on Concordance and Discordance

An essential aspect of medical diagnostic testing using biomarkers is to find an optimal cut-point that categorizes a patient as diseased or healthy. This aspect can be extended to the diseases which can be classified into more than two classes. For diseases with general k ( ) classes, well-established measures include hypervolume under the manifold (HUM), the generalized Youden Index (GYI), and recently proposed maximum absolute determinant (MADET). This research proposes a new measure of diagnostic accuracy, named CD measure that is based on concordance and discordance, for diseases with k classes. The CD measure uses both the correct and incorrect classification information and aims to achieve purposive higher correct classification rates. Moreover, maximizing CD is proposed to be a criterion for selecting optimal cut-points. Simulations for power studies suggest that CD can detect the differences between the null and alternative hypotheses that other methods cannot for some scenarios. Simulation results indicate that using CD to select optimal cut-points can provide relatively high total correct classification rates with less loss than MADET, MD, and MV, and more balanced correct classification rates than GYI. Furthermore, CD measure, along with other methods for comparison, is applied to the ANDI data to choose biomarkers for the diagnosis of Alzheimer’s Disease (AD) and to select optimal cut-points for the chosen biomarkers.