This paper presents a machine learning pipeline designed to predict success in the “Introduction to Clinical Databases” course within the second-year undergraduate degree in Biomedical Engineering. The course encompasses topics related to clinical data, database management systems, and addresses specific challenges inherent to clinical data such as privacy and healthcare regulations compliance. The pipeline involves dataset creation, exploratory analysis, feature selection, data imputation, normalization, regression modeling, and evaluation. A dataset was collected from 30 students, and various regression models were applied to predict student performance using metrics like Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared (R2). Linear Regression (LR) and Random Forest Regression (RFR), using a subset of 20 features, consistently outperformed other models across different target variables. Specifically, for predicting Quizzes grades, 1.4 MAE, 3.0 MSE, and 0.28 R2 were achieved with LR. For predicting the grade in the Project, 1.3, 2.6, and 0.3 were attained for MAE, MSE, and R2, respectively, also with LR. For the projection of Exam grades, LR accomplished 1.4 MAE, 2.9 MSE, and 0.2 R2. Finally, for the Final Grade, the values for MAE, MSE, and R2 were 1.4, 2.9, and 0.2 for LR.