This chapter provides a detailed review of performance evaluation metrics essential for assessing artificial intelligence (AI) algorithms. It emphasizes the importance of choosing appropriate metrics to ensure reliability, generalizability, and fairness in AI model performance. Key evaluation parameters, including accuracy, sensitivity, specificity, precision, recall, F1 score, and area under the receiver operating characteristic curve (AUC-ROC), are explained in the context of both classification and regression tasks. The chapter also discusses the implications of overfitting and underfitting, guiding readers on how to balance model complexity with data variability. A structured, step-by-step framework for machine learning (ML) analysis is presented, highlighting data preparation, model selection, validation, and optimization. By understanding these metrics and evaluation strategies, researchers and clinicians can better interpret AI model outcomes and improve their deployment in healthcare and ophthalmology.