ARTIFICIAL INTELLIGENCE - MACHINE LEARNING - BIG DATA Single channel
Chair (Coordinator) and Rapporteur: GIORGIO GRANI
Module 1: Internal Medicine
- Activity type
- Fisiopatologia, metodologia clinica, propedeutica clinica e sistematica medico-chirurgica
- SSD
- MED/09
- Year
- N/D
- Semester
- N/D
- CFU
- 1
- Hours distribution
- 13 classroom hours
- Lecturers
- GIORGIO GRANI
Module 2: Gastroenterology
- Activity type
- Funzioni biologiche integrate di organi, sistemi e apparati umani
- SSD
- ING-INF/05
- Year
- N/D
- Semester
- N/D
- CFU
- 8
- Hours distribution
- 104 classroom hours
- Lecturers
- PIETRO LIO
Module 3: General surgery
- Activity type
- Fisiopatologia, metodologia clinica, propedeutica clinica e sistematica medico-chirurgica
- SSD
- MED/18
- Year
- N/D
- Semester
- N/D
- CFU
- 1
- Hours distribution
- 13 classroom hours
- Lecturers
- ANDREA POLISTENA
Module 4: Artificial intelligence - machine learning - big data
- Activity type
- Fisiopatologia, metodologia clinica, propedeutica clinica e sistematica medico-chirurgica
- SSD
- MED/12
- Year
- N/D
- Semester
- N/D
- CFU
- 1
- Hours distribution
- 13 classroom hours
- Lecturers
- FLAMINIA FERRI
Learning outcomes
The main goal of the course is to provide students with a solid understanding of the fundamental concepts and key techniques of Machine Learning.
Learn the theoretical foundations and main techniques for collecting data of interest to healthcare.
Be able to evaluate an artificial intelligence model for medical use for diagnostic purposes.
Be able to evaluate an artificial intelligence model for medical use for prognostic purposes.
Understanding the limitations of AI systems in medicine.
At the end of the course, students will be able to:
Distinguish and explain the concepts of supervised, unsupervised, and (broadly) reinforcement learning.
Understand the mathematical and algorithmic principles underlying the main ML models covered.
Implement ML algorithms using standard libraries (e.g., Python with Scikit-learn).
Apply clustering techniques (K-Means, Hierarchical, Lovain) and dimensionality reduction (PCA).
Build and evaluate predictive models (Logistic Regression, Trees, Forests, XGBoost).
Understand the basic concepts of neural networks (CNNs, Autoencoders) and Transformers.
Select the most suitable ML approach based on the problem and available data.
Critically evaluate the performance and interpretability of models.
Work with real datasets using appropriate tools.
Upon completion of the course, students should be able to select, implement, and evaluate appropriate ML algorithms to solve specific problems, understanding their strengths, weaknesses, and interpretability.
Prerequisites
At the beginning of the course, you will be asked to complete a short online form to gather information about your expectations and background, in order to best tailor the learning experience.
Recommended Prerequisites:
Basic programming knowledge (preferably Python).
Basic concepts of probability and statistics.
Familiarity with data analysis is an advantage.
Programme
Module: Internal Medicine
Applications of artificial intelligence in medicine: for diagnosis, prognosis, prediction, therapy.
Sources of big data of healthcare interest: electronic medical records, administrative data, hospital discharge forms, -omics technologies (genomics, transcriptomics, proteomics, multiomics)
Digital Health and e-Health as "efficient and secure use of information and communication technologies to support health and health-related sectors, including health care, health surveillance and health education, knowledge and research".
Wearable devices and the "Internet-of-things".
Data sharing and security.
The contribution of big data and artificial intelligence to precision medicine and the medicine of the four Ps (preventive, predictive, personalized and participatory).
Potential limitations: overfitting, spurious correlations, "black boxes"
Module: Gastroenterology
The course will explore the main paradigms of machine learning: supervised and unsupervised. Both methods considered more easily interpretable and more complex techniques will be covered.
Supervised Learning (Interpretable):Logistic Regression, Decision Trees, Random Forest, XGBoost. We will analyze how these methods handle many variables and provide their significance, with a focus on Logistic Regression for binary outcomes. Outcome evaluation metrics.
Unsupervised Learning (Interpretable):K-Means Clustering (with focus on cluster number input), Hierarchical Clustering (including phylogenetic trees), Lovain Clustering (for complex networks, based on communities and modularity), Network Medicine (applications from omics to clinical and social data), PCA (Principal Component Analysis for dimensionality reduction).
Supervised Learning (Difficult to Interpret):Convolutional Neural Networks (CNNs), CNNs for Biomedical Images: U-NET, Semantic Image Segmentation, Transformers, Vision Transformers; From Transformers to Language Models, Examples of Language Models in Medicine.What is Explainability (XAI): riconoscere bias. Shap (SHapley Additive exPlanations).
Unsupervised Learning (Difficult to Interpret):Autoencoders, Variational Autoencoders (VAE). Examples of Omics data integration and image analysis.
Other Topics:A brief introduction to Deep Reinforcement Learning will be provided: what an agent is, what rewards and policies are, and an example of its application in medicine.
The course will have a strong practical orientation, using real-world datasets from sources such as UK Biobank, Kaggle Datasets, and Physionet, and relying on common tools and libraries (Tensorflow and Pytorch) and program repositories (Huggingface).
Module: General surgery
Applications of Big Data and Machine Learning to breast pathology, surgically relevant endocrine diseases of the neck, diseases of the biliary tract and pancreas, and surgical pathology of the abdominal wall.
Module: Artificial intelligence - machine learning - big data
AI in Gastroenterology Overview, Use of AI in IBD, acute and chronic hepatitis, HCC diagnosis and management, and liver transplant
Books
Module: Internal Medicine
Materials on the University Moodle platform.
Module: Gastroenterology
Teaching Material:Lecture slides, code notebooks, and relevant scientific articles will be made available on the course's e-learning platform (Sapienza Moodle Elearning Platform).
Module: General surgery
ISBN: 9788821446917
Titolo: Sabiston - Trattato di chirurgia
Autori: Sabiston - Townsend - Beauchamp - Evers - Mattox
Editore: Elsevier - Masson EDIZIONI
Volume: Unico
Edizione: XX 2019
Module: Artificial intelligence - machine learning - big data
Materials on University Moodle platform
Bibliography
Module: Internal Medicine
N/D
Module: Gastroenterology
Recommended Texts (Optional):
Ian Goodfellow, Yoshua Bengio, and Aaron Courville. Deep Learning. MIT Press (free download from the official website)https://www.deeplearningbook.org/
Christopher M. Bishop, Hugh Bishop Deep Learning: Foundation and Concepts, Springer 2024
Aurelien GeronHands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, 3rd edition.
Module: General surgery
N/D
Module: Artificial intelligence - machine learning - big data
N/D
Lessons mode
Teaching Methodologies: Lectures with slides, practical coding sessions (individual or group), guided discussions, analysis of scientific articles and case studies.
Participation: Active participation in lectures and discussions is strongly encouraged. Practical sessions will require the use of your own laptop (or access to lab workstations, if available) with a development environment configured (details will be provided). Individual preparation before lectures (reading materials, reviewing concepts) is essential for effective participation.
Frequency
Attendance:
Mandatory, with registration.
Exam mode
The final evaluation will be based on the following components:
Final Written Exam: 3/4 theoretical questions, short-answer exercises (half an A4 page) + 30 multiple-choice questions.
The final grade will be expressed in thirtieths, by converting the score obtained in the written test.
Example exam questions
The questions are related to the lecture topics: for example,
- definition of Big Data
- diagnostic workup of breast neoplasms
- potential iatrogenic injuries during a total thyroidectomy
- diagnostic pathway for obstructive jaundice
Arguments
Module: Internal Medicine
- Data in clinical medicine
- Data source for healthcare
- Examples: AI for thyroid nodule classification
- Examples: AI for risk stratification of thyroid cancers
- Application examples: challenges for improving the quality of care in oncology
- Knowing how to interpret a predictive model in a clinical setting: rules and checklist
- Overreliance
Module: Gastroenterology
- Supervised Learning (Interpretable)
- Unsupervised Learning (Interpretable)
- Supervised Learning (Difficult to Interpret)
- Unsupervised Learning (Difficult to Interpret)
- Deep Reinforcement Learning
Module: General surgery
- breast pathology
- surgically relevant endocrine diseases of the neck
- diseases of the biliary tract
- diseases of the pancreas
- surgical pathology of the abdominal wall.
Module: Artificial intelligence - machine learning - big data
- AI in Gastroenterology: Overview
- AI applications for inflammatory bowel disease
- Integration of AI into digestive endoscopy techniques
Sustainability goals
- Academic year2026/2027
- Degree program to which the course belongsMedicine and Surgery HT
- Mandatory presenceYes
- Languageita
- CFU11 CFU, distributed among 4 integrated didactic modules
- Total duration143 hours