Class Data Science for Healthcare

Data Science for Health belongs to the bioengineering area and addresses statistical, computational, and artificial intelligence methods applied to biomedical and clinical data. The curricular unit covers data preparation and cleaning, descriptive and inferential statistics, visualization, regression, supervised and unsupervised learning, time series, deep learning, genomics, and bioinformatics. It also addresses model validation, communication of results, privacy, security, bias, and algorithmic fairness. Its relevance within the study programme derives from the growing use of data in health research, prevention, diagnosis, prognosis, and decision support. The unit prepares students to develop and evaluate reproducible, ethically responsible, and clinically contextualized solutions.

Not applicable

1. Introduction to data science in biomedicine: scope, applications, challenges, technologies, and tools. 2. Preprocessing: data import, export, cleaning, transformation, missing data, and outliers. 3. Biomedical statistics: descriptive statistics, distributions, inference, hypothesis testing, ANOVA, t and chi-square tests, correlation, linear and logistic regression, model evaluation, and interpretation. 4. Visualization: heatmaps, scatter plots, histograms, libraries, and dashboards. 5. Machine learning: decision trees, support vector machines, neural networks, time series, clustering, PCA, and t-SNE. 6. Genomics, bioinformatics, and molecular pattern identification. 7. Deep learning: convolutional and recurrent networks and applications in medical imaging. 8. Ethics: privacy, security, bias, algorithmic fairness, and regulation.

Teaching combines flipped-classroom activities, interactive lectures, problem-oriented programming, scientific article analysis, and projects based on biomedical data. Materials are provided in advance to promote prior study and active participation. In theoretical-practical sessions, students use computational tools to import, clean, explore, visualize, and model data, building reproducible pipelines. Health cases guide the selection of statistical methods and algorithms, the definition of metrics, and the clinical interpretation of results. The project integrates source code, documentation, and oral discussion, supporting project-based learning. Individual and collaborative work, formative feedback, and the analysis of bias, privacy, and limitations strengthen autonomy, methodological rigour, scientific communication, and ethical responsibility.

Denny, J. C., & Cimino, J. J. (2019). Biomedical informatics: Computer applications in health care and biomedicine. Springer. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press. Witten, I. H., Frank, E., Hall, M. A., & Pal, C. J. (2016). Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann.