Accurate differentiation between diseased and non-diseased states is vital in clinical diagnostics, with optimal cut-off points crucial for precise classification. This study introduces Matthews Correlation Coefficient (MCC) as a robust metric for evaluating diagnostic accuracy. Unlike traditional measures, MCC accounts for all elements of the confusion matrix—true positives, false positives, true negatives, and false negatives—providing a comprehensive assessment of classification performance. Notably, MCC remains effective in imbalanced class distributions, ensuring a balanced evaluation of true negatives and overall diagnostic reliability. Its ability to offer a more informative predictive performance measure makes it valuable for assessing diagnostic tests across varying prevalence levels. Simulation results demonstrate MCC’s superior performance, often surpassing established methods. Applying the MCC measure and cut-off point selection criterion to real-life data further validates its effectiveness in achieving balanced diagnostic accuracy. MCC frequently outperforms traditional metrics, making it a compelling tool in diagnostic test evaluation.
