This study, presented by Agrawal et al., examined the application of machine learning models on explainable AI (XAI) detecting diseases such as diabetes, anaemia, thalassemia, heart disease, and thrombocytopenia. Using the public dataset library, Kaggle, the researchers compiled blood test reports containing 25 relevant markers namely, hemoglobin, cholesterol, white blood count, to name a few. SMOTE (Synthetic Minority Over-sampling Technique) was used to balance the data such that no dominant group would cause bias in the machine learning algorithm. XGBoost, Naïve Bayes, Decision Tree, and Random Forest were algorithms tested to evaluate their diagnostic abilities while SHAP and LIME were used to apply an explanation for conclusions, thus building the trust of physicians in AI through reasoning. XGBoost outperformed all other machine learning models with an accuracy reaching approximately 99.2% and better precision and F-1 scores compared to the rest. Agrawal et. al acknowledged the limited generalizability of the study given that the ML models were only tested on one dataset and noted the advantage of direct computational links of patient lab results to the proposed diagnosis in question.