Fundamentals of Artificial Intelligence for Data Analysis in Molecular Epidemiology

The course aims to provide students with the theoretical and methodological foundations of Artificial Intelligence applied to the analysis of complex data in molecular epidemiology, with particular emphasis on omics, environmental, clinical, and longitudinal data.
The course is designed to develop interdisciplinary competencies in machine learning, advanced statistical modeling, and multimodal integration of biomedical data, introducing students to the main approaches used to extract knowledge, identify biological patterns, and build interpretable predictive models within the contexts of health biology and personalized preventive medicine.
Particular attention will be devoted to:
· the fundamental principles of supervised and unsupervised learning;
· challenges related to the high dimensionality of biological data;
· integrated analysis of multi-omics and environmental data;
· evaluation of model performance;
· issues of interpretability, robustness, causality, and reproducibility;
· ethical and regulatory aspects of AI applications in biomedicine.
The course also aims to provide students with conceptual and practical tools to critically understand the role of AI in modern epidemiological research and in future translational applications.

At the end of the course, students will be able to:
1. Understand the fundamental principles of Artificial Intelligence and Machine Learning applied to biomedical and epidemiological data.
2. Describe the main types of data used in molecular epidemiology, including genomic, epigenomic, transcriptomic, clinical, and environmental information.
3. Apply preprocessing, normalization, dimensionality reduction, and data integration methods for complex high-dimensional datasets.
4. Implement and evaluate supervised and unsupervised learning models for classification, regression, clustering, and risk stratification problems.
5. Critically interpret model performance using appropriate statistical metrics and validation procedures.
6. Understand issues related to overfitting, bias, confounding, and generalizability in AI models applied to biomedical research.
7. Use computational tools and programming languages for data analysis and the development of reproducible workflows.
8. Critically discuss limitations, ethical implications, and potential applications of AI in predictive and preventive biomedicine.
9. Communicate analytical results and biological interpretations in a rigorous and interdisciplinary manner.