This course presents key health data sources that, while not strictly medical data, can be effectively used for healthcare organization and health science research. Participants will learn about the main types of data sources, methods of data collection, and the biases that may arise during measurement and collection. Ethical considerations related to data collection and analysis will also be addressed. Practical exercises will explore potential data sources, common challenges associated with them, and the application of selected analytical tools in practice, including a brief introduction to data visualization using Power BI.

This short course, “Deep Learning for Unstructured Medical Data,” explores advanced machine learning techniques for analyzing complex healthcare data such as medical images, clinical notes, and genomic sequences. Participants will learn key deep learning architectures, including convolutional neural networks (CNNs) for image analysis and recurrent neural networks (RNNs) and transformers for natural language processing (NLP) in medical contexts. The course emphasizes real-world applications, such as disease diagnosis, patient outcome prediction, and personalized treatment planning. Hands-on sessions using popular frameworks like TensorFlow and Keras will equip learners with practical skills to handle unstructured data effectively.

This lecture series introduces the core concepts of machine learning with a focus on biomedical applications. Theoretical sessions cover fundamental models, including linear and logistic regression, and an introduction to neural networks. Practical sessions provide hands-on experience implementing these models using Python and relevant libraries (scikit-learn, PyTorch, PyTorch Geometric). Participants will work with real biomedical datasets, gaining insights into model training, evaluation, and interpretation. An additional session explores graph neural networks and their applications in biomedical research.

The course aims to provide participants with the skills to analyze, model, and predict the behavior of complex networks. They will explore fundamental concepts of graphs and networks, mathematical representations, and key metrics such as closeness, betweenness, and PageRank. The curriculum covers universal properties of networks, including the small-world effect, scale-freeness, and clustering, as well as foundational network models such as the Erdős–Rényi and Barabási–Albert models. Participants will study network robustness, percolation transitions, and the dynamics of information and epidemic spreading. The course also introduces network motifs and community detection methods to uncover hidden structures, with a particular emphasis on biomedical networks and applications. Theoretical knowledge is reinforced through practical analyses and modeling exercises.