Data Analytics and Artificial Intelligence Health Sciences

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Describe fundamental programming strategies.

Describe the functioning principles of database systems.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Develop advanced object-oriented Python code

Develop code for algorithmic problem solving.

Organize and Use databases.

MISSION
The goal of the course is to provide the students with a wide range of skills and knowledge in computation and data management/analysis, which are extensively used throughout the whole education program.
CONTENT SUMMARY

Python programming: basic and advanced programming techniques.

Python scientific modules

Object-oriented programming.

Relational databases

Other advanced topics

Oral individual exam [General exam]
Written individual exam (traditional/online) [General exam]
Individual Works/ Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
ATTENDING AND NOT ATTENDING STUDENTS
There are no partial exams. The exam consists in 4 programming exercises and some theory questions. The submitted solutions will be discussed during the subsequent oral part. In the oral examination students are required to clearly justify their programming choices.
Failure to provide justification will result in a (possibly heavy) reduction of the final grade.

During the course some assignments will be given to the class. If the assignments' overall score is positive, students can choose to skip one of the four exercises in the exam paper.

All assignments and exams consist of programming exercises and/or questions of several types (multiple choice, true/false, open-ended) whose aim is to assess the understanding of fundamental programming strategies and the knowledge of database systems. Collectively, the exercises will test the ability of students to develop advanced Python code for algorithmic problem solving.

ATTENDING AND NOT ATTENDING STUDENTS
Lecture notes provided by the instructors.

Allen Downey
,
Think Python
3rd edition, available at https://allendowney.github.io/ThinkPython/

Numpy and Pandas documentation (online)

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Recognize the key guiding principles and concepts of statistical inference.

Identify and describe a toolkit of methods and models.

Summarize and defend concepts and knowledge useful in preparation of the courses on Machine Learning and Artificial Intelligence.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Formulate suitable probabilistic models for statistical problems in the health sciences.

Apply appropriate statistical methodology to estimation and testing problems.

Interpret the outcome of the estimation procedures in view of the application at hand.

MISSION
This course will provide the foundational elements of statistical theory, methods and models that are useful in the health sciences. The principles discussed will form the backbone for later developments in the ML and AI courses. Problems of estimation will be dealt with within the finite sample size setting and the large sample size setting. The R software will be used to discuss and appreciate the properties of the procedures, and to implement the methods.
CONTENT SUMMARY
The topics that will be discussed are:

Review of frequentist Inference. Joint and conditional distributions, Independence, point estimation, Information, Confidence intervals; Hypothesis testing.

Power functions and p-values.

Large-sample approximation and convergences.

Maximum likelihood estimators. Newton-Raphson algorithm.

The multivariate normal distribution. The linear model. General linear hypotheses.

Simultaneous confidence intervals. Bonferroni.

ANOVA.

Sample size determination: two-group testing.

Binary data. Odds ratios. Stratification, homogenous association, conditional independence. Inference.

Regression for binary outcomes: logistic regression.

Classification. Screening tests. ROC and AUC.

Intro to Survival Analysis: Kaplan-Meier, Log-rank test, Modified Wilcoxon test.

Content of the labs:

1.
Introduction to R

2.
Statistical Inference

3.
Maximum Likelihood Estimation (MLE) and Introduction to the Linear Model

4.
The Linear Model and ANOVA

5.
Logistic Regression

6.
Classification and Survival Analysis

Written individual exam (traditional/online) [General exam]
ATTENDING AND NOT ATTENDING STUDENTS

A 2-hour in-class, closed book and closed notes exam will be held after the course, within each of the three exam sessions.

The exam will be handwritten on paper.

Some of the questions (for a total of 21 points out of 31) will cover the methods, models, and properties discussed in class, to assess the understanding of the theoretical aspects.

Some of the questions (for a total of 10 points out of 31) will require the analysis of some data using R. Students will bring their own laptop to the exam and will not be required to submit any of the code or output: only the final answers of the analyses and the associated comments will need to be reported on the paper. No access to the internet will be allowed during the exam.

There will be no difference in assessment method / exam program between attending and non attending students.

NOTE: While there will be no difference in assessment method / course content between attending and non attending students,
attendance is strongly recommended
.

ATTENDING AND NOT ATTENDING STUDENTS
Materials for the course will be distributed through the Blackboard platform. They will consist of class notes, R code, and additional handouts.

For more details on the topics we suggest the following volumes:

Dalgaard P (2008). Introductory Statistics with R. Springer, New York, NY.

Dobson AJ and Barnett AG (2018). An Introduction to Generalized Linear Models, Fourth Edition. Chapman & Hall / CRC, Boca Raton, FL.

Lawless J (2003). Statistical Models and Methods for Lifetime Data. Wiley, New York, NY.

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Describe the theoretical foundations of artificial neural networks and their learning dynamics

Explain the role of regularisation techniques and weight initialisation in deep model training

Illustrate the effects of overparameterisation, including the bias-variance trade-off and the double descent phenomenon

Distinguish between major deep learning architectures, such as CNNs, RNNs, transformers, and graph neural networks

Summarise the principles and use cases of generative models

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Identify appropriate architectures for different types of biomedical data (e.g., images, sequences, graphs) and apply them to analyse and interpret the data

Implement neural networks tailored to specific biomedical problems using Python and PyTorch, and evaluate their performance

Analyse the training behaviour of neural networks and interpret performance metrics to guide model refinement

Compare different deep learning approaches (e.g., CNNs, RNNs, Transformers, GNNs, VAEs, GANs) in terms of suitability for a given biomedical task

Recognise current limitations and emerging research challenges in AI applied to biomedicine

Collaborate effectively in small teams to develop and present solutions to real-world biomedical challenges

Communicate methodological choices and results clearly and effectively through oral presentations and written technical reports

MISSION
This course deepens the theoretical and practical understanding of artificial neural networks and their applications in biomedical domains. It introduces key concepts behind modern deep learning architectures, focusing on how these systems learn, generalise to unseen data, and generate new data. Students will explore most of the existing neural network classes, including convolutional, recurrent, and transformer-based architectures, with an emphasis on biomedical data, including such multi-omics data and images. The course also covers regularisation strategies, weight initialisation, and the effects of overparameterisation, providing insights into the bias-variance trade-off and the double descent phenomenon. In the second part, students will be introduced to unsupervised learning, focusing on deep generative models, including Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), and their applications in biomedical research. Practical sessions in Python, using PyTorch and related tools, will enable students to implement, train, and evaluate their models on real-world biomedical datasets.
CONTENT SUMMARY

Theoretical foundations of artificial neural networks and backpropagation

Regularisation techniques (e.g., weight decay, dropout, early stopping)

Initialisation strategies and their impact on training dynamics

Overparameterization, bias-variance trade-off, and the double descent phenomenon

Convolutional Neural Networks (CNNs) and their application to biomedical image and video data

Recurrent Neural Networks (RNNs) and their applications to temporal and sequential biomedical data

Transformer architectures and attention mechanisms in life sciences

Graph Neural Networks (GNNs) for structured biomedical data

Unsupervised learning with deep models: Autoencoders, Variational Autoencoders (VAEs), and Generative Adversarial Networks (GANs)

Hands-on labs with real-world biomedical datasets

Current challenges and future perspectives in AI for biomedicine

Oral individual exam [General exam]
Written individual exam (traditional/online) [General exam]
Individual Works/ Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
Collaborative Works / Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
ATTENDING STUDENTS
Students who attend at least 75% of the course activities are considered attending students. In addition to the written exam, attending students will participate in a group project related to the course content. The project involves designing and implementing deep learning models for biomedical data, culminating in a final group presentation. The project aims to evaluate the "applying knowledge and understanding' learning objectives. The
final grade
will be determined as the average of the written exam (50%) and the group project (50%). A minimum passing grade of
16 out of 31
must be achieved in
both parts
. The final grade, rounded up, must be at least
18
to pass the course.
NOT ATTENDING STUDENTS
Students who do not meet the minimum attendance requirement (75%) must take the
general final exam
. In place of the group project, non-attending students will do an
individual project
, focused on a topic proposed or approved by the instructors. The project involves the independent development and implementation of deep learning models for the chosen topic, accompanied by a
written final report
.

ATTENDING AND NOT ATTENDING STUDENTS

Bishop, Chris, and Bishop, Hugh.
Deep Learning: Foundations and Concepts,
Springer, 2024

Prince, Simon J.D.
Understanding Deep Learning,
MIT Press, 2023

Scardapane, Simone.
Alice's Adventures in a Differentiable Wonderland (Volume I, A Tour of the Land), 2024

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Provide an overview of the potential outcomes approach for causal inference.

Understand the basic assumptions underlying causal analysis.

Master statistical methods for learning causal relationships.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Identify causal inference problems in health applications.

Analyze causal inference problems and identify the most suitable research design for addressing them, both with experimental and observational data.

Apply causal inference methods to real-world data, either experimental or observational, with the aid of software tools.

MISSION
While introductory statistics courses develop a multivariable thinking approach that relies on well-known association measures for studying dependence, in health and biomedical applications, one is more often interested in the causal effect of an intervention, or treatment, on an outcome. Lectures will provide both a methodological and a hands-on perspective on several problems, wherein the ultimate goal is to link data analysis to causal conclusions. The course will start with the simpler setting of randomized experiments, before addressing more complex situations where causal relationships are inferred from observational data. Practical examples based on public health applications will be developed throughout.
CONTENT SUMMARY

Review of measures of association. Contingency tables, risk difference, risk ratio, odds ratio, Simpson’s paradox.

Basic ideas of causal inference as a missing data problem. The potential outcomes framework. Counterfactuals and causal effects. Average treatment effects.

Randomized experiments: Fisher’s exact test and Neyman’s approach to repeated sampling.

Observational studies. Unconfoundedness and propensity scores.

Regression discontinuity designs.

Causal survival analysis.

Unmeasured confounding and instrumental variables methods.

Bayesian causal inference.

Applications in health.

Written individual exam (traditional/online) [General exam]
ATTENDING AND NOT ATTENDING STUDENTS
For determining the final assessment, students must choose between the following two options

1)
The general exam, at the end of the course.

2)
The combination of

2.1)
A group project presentation or a group paper presentation, accounting for 40% of the final grade. Such presentations will take place during the course.

2.2)
The general exam, accounting for 60% of the final grade, at the end of the course.

As for 2.1), each group is identified by the instructors and must express their preference over presenting a paper versus a project.

Project:
a group will be assigned a dataset for which a causal analysis must be conducted using the tools that have been discussed during the course. The causal analysis and the results must be presented, with slides, and students must point out connections with the material presented during the course.

Paper:
a group will be assigned a research paper on causality and will have to deliver a presentation discussing it, with slides. The presentation should summarize the main findings of the paper and emphasize the connections with the topics discussed during the course.

General exam:
it consists of a written test, with questions on both the methodological and the applied parts of the course, and aims at testing students’ understanding of the statistical foundations and the practical uses of modern approaches to causal inference problems.

ATTENDING AND NOT ATTENDING STUDENTS

Ding, P. (2024).
A first course in causal inference
. CRC Press. A preliminary draft of the textbook is available at https://arxiv.org/pdf/2305.18793

Additional references will be provided during the course.

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
Define what is a program in Python

Describe the grammars of Python
APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
Correctly write programs in Python

MISSION
This preparatory course introduces students to the basic programming techniques in Python.
CONTENT SUMMARY
Python Language:

Basics

Data Types, Strings, Collections, The Standard Library

Control Flow

Functions

Modules

OOP

Input/Output

Exceptions, Iterators and Generators

Functional Programming

Numpy, Pandas

Active class participation (virtual, attendance) [Continuous assessment]
ATTENDING AND NOT ATTENDING STUDENTS
Being a preparatory course, there is no assessment.

ATTENDING AND NOT ATTENDING STUDENTS
Think Python, 3rd Edition (
online version available for free
) by Allen B. Downey O'Reilly Media

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Understand the organization and functioning of healthcare systems, including their data flows, regulatory principles, and governance structures.

Recognize the different types of health data — clinical, administrative, epidemiological, imaging, and omics — and their relevance for research, management, and policy design.

Analyze how data systems and digital infrastructures support healthcare delivery, performance evaluation, and innovation in clinical and policy contexts.

Evaluate key challenges in health data governance, including privacy, interoperability, ethical considerations, and the regulatory frameworks that shape data use and sharing.

Apply conceptual frameworks to assess how data and information systems can enhance evidence-based decision-making and the overall efficiency of healthcare systems

Develop actionable recommendations for data governance and policy design in realistic healthcare scenarios.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Analyze how healthcare data flow across organizations and how system design, governance, and regulation shape their use.

Evaluate the strengths, limitations, and biases of different types of health data (e.g., clinical, administrative, imaging, digital).

Assess the implications of data governance, privacy, and interoperability for real-world decision-making in healthcare.

Apply economic evaluation methods (e.g., CEA, CBA, HTA) to compare health technologies and inform policy or managerial choices.

Interpret measures of value—such as QALYs, patient preferences, wellbeing indicators—and use them to support evidence-based recommendations.

Construct simplified decision models (e.g., decision trees, Markov models, budget impact analyses) to forecast health and economic outcomes.

Collaborate effectively in teams to solve simulated policy or management challenges, integrating diverse perspectives and responsibilities.

Communicate complex data insights clearly and persuasively in presentations, discussions, and written assignments.

Reflect critically on the societal, ethical, and organizational implications of data use in healthcare systems.

MISSION
The mission of Data Systems in Healthcare is to empower future health leaders with the ability to interpret, question, and strategically use data to transform healthcare systems. The course seeks to cultivate a generation of professionals who can bridge technical literacy with policy insight, turning complex data environments into actionable knowledge that advances patient wellbeing, organizational performance, and societal impact. By fostering critical thinking, interdisciplinary dialogue, and hands-on analytical competence, the course aims to equip students to shape responsible innovation, design sound governance frameworks, and drive evidence-informed decision-making across health institutions and organisations.
CONTENT SUMMARY
The Data Systems in Healthcare course offers an in-depth exploration of how health systems generate, structure, and leverage data to drive decision-making, innovation, and policy transformation. At the crossroads of health management, data science, and digital innovation, the course equips students with the analytical and critical tools needed to understand and shape the data ecosystems underpinning modern healthcare.

Through a blend of conceptual frameworks, case-based discussions, and hands-on simulations, students will learn how data flow across hospitals, regional systems, and national infrastructures, and how governance models, interoperability standards, and regulatory frameworks influence their use. Real-world examples will illustrate how data—from clinical records and administrative databases to imaging, genomic, and digital health data—can be harnessed to support clinical research, improve patient outcomes, and inform policy choices.

By the end of the course, students will be able to navigate the complexity of healthcare data systems with both technical literacy and strategic awareness, understanding how data governance, ethics, and AI applications can jointly shape the future of sustainable healthcare.

Written individual exam (traditional/online) [Partial exams]
Collaborative Works / Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
Active class participation (virtual, attendance) [Continuous assessment]
ATTENDING STUDENTS
Component
Active Participation and Engagement
Data System Challenge (Team Simulation)
Final Case-Based Exam (Individual)
NOT ATTENDING STUDENTS
Written exam (made of 10 multiple choices in 40 minutes) on all reading materials as specified in the syllabus and made available in the e-learning.

ATTENDING AND NOT ATTENDING STUDENTS
There is no textbook for attending students. All the reading materials will be made available. For detailed instructions refer to the syllabus.

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
At the end of the course student will be able to:

Develop the ability to scrutinize complex systems using modeling and simulation techniques.

Utilize acquired skills in population dynamic modeling, agent-based modeling, and network science for health

Investigate underlying mechanisms in diverse biological and population processes, allowing for a comprehensive exploration of real-world applications.

Develop the capacity to formulate projections for future scenarios based on alternative assumptions.

Acquire innovative tools from AI techniques

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
At the end of the course student will be able to:

Apply advanced modeling and simulation techniques to address and solve complex health-related challenges

Utilize acquired skills in population dynamic modeling, agent-based modeling, and network science to analyze and solve real-world interdisciplinary problems

Integrate modeling and simulation skills into counterfactual scenario analysis and decision making processes

Formulate informed projections for future scenarios by applying alternative assumptions

Evaluate the appropriateness of different modeling techniques and adapt these to diverse health-related problems

Apply soft skills such as effective teamwork, clear communication, and collaborative approach when attending classes

Demonstrate analytical thinking by recognizing patterns and trends in simulated scenarios, leading to the identification of optimal solutions

Communicate insights of modeling and simulation outcomes clearly and effectively to diverse stakeholders in a professional context

Exhibit confidence in the application of modeling and simulation methodologies across public and private sectors, contributing to effective problem-solving and translating theoretical knowledge into practical solutions

MISSION
Mathematical models have been increasingly used to understand infectious disease dynamics within and between populations and, more generally, to study complex systems. This course offers a comprehensive exploration of modeling and simulation techniques within the realm of Data Science. It places emphasis on population dynamic modeling, agent-based modeling (ABM), and the foundational principles of network science. Students will develop crucial skills for analyzing and interpreting intricate human, social and biological systems, enhancing their ability to navigate decision-making processes leveraging on the available data.
CONTENT SUMMARY
The course will cover the following topics:

Discrete-time deterministic and stochastic models

Ordinary differential equation models

Metapopulation models

Simulation, sensitivity and sampling parameter sets, including MCMC techniques for model calibration

Network models: reading adjacency matrices and simulation of Reed-Frost models

Agent-based models

AI extensions of epidemic models

Written individual exam (traditional/online) [General exam]
Individual Works/ Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
ATTENDING AND NOT ATTENDING STUDENTS
Assessment is based on:

A written exam (i.e. 50%):

The exam will be closed book (Respondus) with open and closed questions. The exam covers material covered in the lectures, practical classes, in the text books and other set of readings provided by the Professors. The exam has to be done in one shot at the end of the course.

An individual project (50%):

The project can be submitted once only and will remain valid throughout the entire academic year. The mark will consider both the validity, robustness and clarity of the written report as well as the outcome of the individual oral presentation in terms of the idea, method and results of the project.

ATTENDING AND NOT ATTENDING STUDENTS
Slides will be made available by professors in advance of the lectures and will represent the main guiding reference for the preparation of the exam. Programming codes, reading materials and selected chapters of relevant books will be uploaded on the e-learning platform.

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Design and perform data-driven analyses for interpretation, prediction, and classification, addressing various aspects and subtleties of model selection.

Understand and apply the bias-variance trade-off and shrinkage methods.

Implement resampling methods for reliable model assessment.

Identify and use tree-based and kernel-based algorithms.

Explain principles of large-scale hypothesis testing and control of error rates.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Formalize real-world problems as precise statistical questions.

Make informed judgments by selecting appropriate machine learning tools.

Assess statistical significance and uncertainty of analysis results

Lifelong Learning Skills

By the end of the course, students will be able to:

● Balance the trade-off between model complexity and performance in learning algorithms.

● Develop algorithms to address relevant and evolving learning problems.

● Adapt their analytical skills to ongoing data challenges and emerging technologies in health and biomedical sciences.

MISSION
The course addresses core topics in statistical machine learning that are central to modern data analysis, with a particular focus on their application to health and biomedical sciences. As digitalization in healthcare leads to increasingly large, high-dimensional, and complex datasets, there is a growing need for methods that combine predictive accuracy with interpretability. This course equips students with both theoretical understanding and practical tools to meet this challenge. It covers essential models and algorithms—from classical regression and resampling to tree-based methods, support vector machines, and graphical models— along with their applications in uncertainty quantification and multiple testing. By bridging statistical theory with hands-on implementation in Python, the course provides a solid foundation for students to critically apply machine learning methods to real-world problems in health sciences, and contributes to forming professionals capable of both methodological innovation and responsible application in AI-driven healthcare.
CONTENT SUMMARY
1. An introduction to statistical machine learning. Supervised learning; prediction accuracy vs model interpretability; assessment of model accuracy; methods for evaluating predictive uncertainty.

2. Review of regression and classification.

3. Bias-variance trade-off and shrinkage methods.

4. Resampling methods: cross-validation and bootstrap.

5. Nonlinear models: polynomial regression, regression and smoothing splines, generalized additive models.

6. Tree-based methods and their applications in survival analysis: regression and classification trees; bagging; random forests; support vector machines; kernel methods; boosting.

7. Large-scale hypothesis testing: FWER, Bonferroni, FDR, Benjamini-Hochberg, resampling-based methods.

8. Learning with directed and undirected graphical models. Identification of conditional independencies. Exact and approximate inference methods.

Written individual exam (traditional/online) [General exam]
ATTENDING STUDENTS
The assessment consists of a single final exam (no partial exams), composed of two parts held on the same day:

• Theoretical part: a written component with open-ended questions, designed to test understanding of foundational concepts.
• Applied part: an individual oral component, designed to evaluate practical problem-solving and implementation skills.

NOT ATTENDING STUDENTS
Non-attending students will take the standard final exam and will also be asked one additional oral question. This question may be either theoretical or applied.

ATTENDING AND NOT ATTENDING STUDENTS
The following books and resources will be used as primary and supplementary references:

● James, G., Witten, D., Hastie, T., Tibshirani, R. (2023). An Introduction to Statistical Learning (Python edition). Springer. (freely available in PDF format at https://www.statlearning.com/)

● Hastie, T., Tibshirani, R., Friedman, J. (2009). The Elements of Statistical Learning. Springer. (freely available in PDF format at https://hastie.su.domains/ElemStatLearn/)

● Murphy, K. (2022). Probabilistic Machine Learning: An Introduction. MIT Press. (freely available in PDF format at https://probml.github.io/pml-book/book1.html)

● Højsgaard, S., Edwards, D., & Lauritzen, S. (2012). Graphical models with R. Springer Science & Business Media.

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
Demonstrate basic knowledge of linear algebra.
Linear algebra course covers the following topics: vectors, vector spaces, matrices, linear maps, eigenvalues and eigenvectors, spectral theorem, and singular value decomposition.
APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
Understand the fundamental concepts of linear algebra and probability theory, and solve basic exercises.

MISSION
This preparatory course introduces the basis of linear algebra and probability theory.

In the first part of the course, we will cover some basic topics of linear algebra, including vectors, matrices, linear systems, vector spaces, linear maps, eigenvalues and eigenvectors, the spectral theorem, and the singular value decomposition.
CONTENT SUMMARY

Complex Numbers

Vectors

Linear Systems and Matrices

Vector Spaces

Linear Maps and their Matrix Representation

Invertible Linear Maps and Isomorphism

Norms and Inner Products

Eigenvalues and Eigenvectors

Change of Basis

Spectral Theorem

Positive Definite and Semidefinite Matrices

Active class participation (virtual, attendance) [Continuous assessment]
ATTENDING AND NOT ATTENDING STUDENTS
The course has no exams.

ATTENDING AND NOT ATTENDING STUDENTS
Suggested textbooks:

Sheldon Axler,
Linear Algebra Done Right

Marc Peter Deisenroth, A. Aldo Faisal, Cheng Soon Ong,
Mathematics for Machine Learning

Gilbert Strang,
Introduction to Linear Algebra

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
At the end of the course student will be able to...

Describe different text analysis problems.

Talk about the linguistic foundations.

Distinguish between exploration and prediction approaches.

Know which algorithm to choose for a given problem.

Understand the trade-offs between different approaches.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...
At the end of the course student will be able to...

Apply their knowledge to practical text analysis problems.

Implement a variety of algorithms for text exploration and classification in Python.

MISSION
Natural Language Processing tools are becoming ubiquitous: from everyday tools like chatGPT, Siri, or Alexa to decision making processes in industry or politics and to text analysis tools in social science research. Machine-learning based text analysis tools provide a range of possibilities and are a growing field of expertise. Whether it is the exploration of text to find structures and topics, or the construction of a classifier to predict the sentiment or author characteristics of a text, this course provides an overview and hands-on experience in all relevant techniques.
CONTENT SUMMARY
Preparation: how do I work with text:

Data formats.

Preprocessing.

Storage and retrieval.

Exploration: exploring structure in the data:

Clustering.

Topic models.

Static and contextual word embeddings.

Prediction: finding patterns to impute new values:

Text classification (sentiment analysis, author attributes).

Logistic Regression.

Perceptron.

Feed-forward Neural Nets.

Transformer architecture

BERT

chatGPT and (Large) Language Models

Individual Works/ Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
Collaborative Works / Assignment (report, exercise, presentation, project work etc.) [Continuous assessment]
Active class participation (virtual, attendance) [Continuous assessment]
ATTENDING AND NOT ATTENDING STUDENTS

Individual works: by completing coding exercises or presenting results, the student is encouraged to work independently to tackle conceptual and practical problems.

Collaborative works: by completing exercises or preseting results in groups, student are encouraged to discuss among peers and then converge to a common outcome.

Active class participation: in-class discussions allow the teacher to better modulate the pace for the class, and the students to have difficult points clarified.

ATTENDING AND NOT ATTENDING STUDENTS

Lecture slides and notes provided on Bboard.

D. HOVY,
Text Analysis in Python for Social Scientists: Discovery and Exploration
, Cambridge University Press, 2020. (https://www.cambridge.org/core/elements/text-analysis-in-python-for-social-scientists/BFAB0A3604C7E29F6198EA2F7941DFF3)

JURAFSKY, DAN, J.H. MARTIN,
Speech and language processing, Vol. 3.
London, Pearson, 2014.

C.D. MANNING, H. SCHUTZE,
Foundations of statistical natural language processing,
MIT press, 1999.

S. MARSLAND,
Machine learning: an algorithmic perspective,
CRC press, 2015.

F. CHOLLET,
Deep learning with Python
, Manning Publications Co., 2017.

KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Describe the main principles and structure of the GDPR, including key concepts such as personal data, lawful processing, consent, and data subject rights.

Explain the aims, scope, and major mechanisms of the proposed EU AI Act, with specific emphasis on high-risk AI applications in healthcare.

Identify the ethical challenges and dilemmas related to the use of personal data and AI-based systems in health sciences, including issues of bias, accountability, and transparency.

Recognize the roles of different actors involved in data governance in the health domain, including data controllers, processors, supervisory authorities, and health professionals.

APPLYING KNOWLEDGE AND UNDERSTANDING
At the end of the course student will be able to...

Analyze real-world scenarios in health research and practice to determine compliance with GDPR and AI Act requirements.

Assess potential risks related to privacy and ethics in digital health initiatives and suggest mitigation strategies in line with EU regulatory frameworks.

Apply principles of data protection by design and ethical AI development to hypothetical health-based data projects.

Communicate regulatory and ethical considerations effectively in interdisciplinary teams and to non-specialist audiences, using clear and accessible language.

MISSION
Privacy, ethics, and regulation are foundational concepts in today’s data-driven society, especially in the health sciences. As digital technologies evolve, so do the legal and moral challenges around personal data, artificial intelligence, and automated decision-making. This course introduces to the core principles of European regulatory frameworks, particularly the GDPR and the AI Act, with an emphasis on their application in healthcare and biomedical research.
The goal is to equip students in the Data Analysis for Health Sciences program with the ability to understand and critically evaluate regulatory and ethical issues arising from the use of personal data and AI systems in health-related contexts. By the end of the course, students will have developed interdisciplinary competencies to navigate privacy, compliance, and ethical challenges in digital health innovation.
CONTENT SUMMARY
Part I – EU legislation shaping AI in healthcare

· Framing of the European Health Data Space (EHDS)

· Existing Legislation

· Legal relevance for research and practice

Part II – Fundamentals of EU Legislation on Data Protection and Governance

· GDPR

· Data Governance Act

· Data Act

Part III – The Deployment of AI in Healthcare

· AI Act

· Product Liability Directive

· Case scenarios

Written individual exam (traditional/online) [General exam]
ATTENDING STUDENTS
Group assignment and written individual exam.
NOT ATTENDING STUDENTS
Written individual exam.

ATTENDING AND NOT ATTENDING STUDENTS
Reading and materials uploaded on Bboard for students.