In this work, we present the results obtained after using some of the most modern Machine Learning (ML) algorithms: K-Means (KM), Decision Trees (DT), Logistic Regression (LR), Bernoulli Naive Bayes (BNB), Gaussian Naive Bayes (GNB), Random Forest (RF), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Gradient Boosting (GB). We worked with real clinical data that refer to kidney transplantation. Features include age, gender, Killer-cell immunoglobulin-like receptor (KIR) characteristics and patient status. The conclusion depends on the small volume of data available and tells us that on average, using the above characteristics to classify the patient’s status, the accuracy does not exceed 70% for any of the algorithms we used. GB and GNB outperformed all the other estimators (accuracy 70%) with penalty in execution time for GB—32.5 min, compared to other estimators where the execution time was under 1 min. GNB was the best to precision. SVM was the best to specificity. GB was also the best at recall and F1 score.