LEARNING OUTCOMES

Il corso introdurrà i concetti ei principi fondamentali dell'apprendimento automatico e dell'intelligenza artificiale applicata alla medicina

PREREQUISITES

Recommended background knowledge

Linear algebra
Probability and basic statistics
Calculus 

TEACHING METHODS

Duration of teaching 48 hours (1/2 theory 1/2 practice).

• Classes taught in person by the lecturer with the use of slides and/or blackboard.
• Hands-on practice using Jupyter notebooks.

SYLLABUS/CONTENT

Introduction to Artificial Intelligence. Examples of AI in medicine and biology. Data driven methods: supervised /unsupervised. Supervised: classification/regression.Unsupervised: clustering and visualization. Bias and Generalization  

Machine learning methods for prediction. Data: issues and challenges. Methods: examples of predictive models for diagnosis or prognosis. Detection of disease. Disease progression prediction. Disease staging. Regularization methods. Measures to assess the classification model

Deep learning methods for prediction. Examples of predictive methods for diagnosis or prognosis. Early detection of disease. Disease progression prediction. Disease staging. Deep learning. Multilayer perceptron. Model dependency on learnable parameter and model stochasticity

Applications of AI in medicine. Medical image prediction. Deep learning (CNN)​. Graph neural networks

Examples of variable selection in medicine. Case study: identification of pathogenic molecular variables. SVM-RFE. Lasso and sparsity inducing methods.Variable selection for genomics

RECOMMENDED READING/BIBLIOGRAPHY

• Hastie, Trevor, et al. The elements of statistical learning: data mining, inference, and prediction. Vol. 2. New York: springer, 2009.
• Hastie, Trevor, Robert Tibshirani, and Martin Wainwright. "Statistical learning with sparsity." Monographs on statistics and applied probability 143 (2015): 143.
• Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning. MIT press, 2016.