The Machine Learning for Healthcare track (closely linked to the Medical ICT one, to be found here) provides competences to apply machine learning to medical care scenarios.
Fields of special interest are medical networking applications, signal processing for bioimaging, neurosciences, eHealth/mHealth, physical rehabilitation, and every other field where machine learning strategies can lead to an improvement in human quality of life.
A list of the courses available in the “Machine Learning for Healthcare” curriculum for the 2022-2023 academic year is reported below.
The mandatory courses are Signal Processing for Life and Health (that includes Digital Signal Processing and Machine Learning) and Human Data Analytics.
The other courses can be freely chosen subject to some ground rules: at least 5 courses from the ICT subjects and 3 courses from the related subjects must be chosen.
Furthermore, a soft skill course must be selected, and students must pass the B2 English exam.
It is further advised to choose about 30 ECTS credits per semester to keep your workload balanced.
The Signal Processing for Life and Health course is mandatory, and composed of 2 parts: Digital Signal Processing and Machine Learning.
Furthermore, the Human Data Analytics course must also be mandatorily taken.
The course exploits basic signal analysis knowledge that the student is assumed to have acquired from previous studies to explore advanced concepts in the field of digital signal processing.
The course will review Z-transform, linear time-invariant systems, FIR/IIR filters, to investigate the design and usage of digital filters, interpolation/decimation of digital signals, frequency analysis of digital signals.
Practical application examples, useful in many areas of information engineering, will be provided.
Intelligent systems capable of automated reasoning are emerging as the most promising application of ICT.
The aim of this course is to provide fundamentals and basic principles of the machine learning problem as well as to introduce the most common techniques for regression and classification.
Both supervised and unsupervised learning will be covered, with a brief outlook into more advanced topics such as Support Vector Machines, neural networks and deep learning.
The course will be complemented by hands-on experience with Python programming.
The course deals with the analysis of biosignals, i.e., signals generated by human activity.
It will study extraction of information from quasi-periodic signals (denoising, segmentation); application of clustering algorithms over biosignals to classify users and construct dictionaries for compact biometric datasets; use of unsupervised learning to perform quantization; statistical structures such as Bayesian networks and Hidden Markov Models, as well as supervised learning, for pattern and classification problems.
Select 5 courses from the following list
The course offers a guided tour of 3D computer vision, 3D graphics and machine learning tools to develop virtual and augmented reality applications.
After a description of imaging systems, the course reviews how to build a 3D model starting from 2D pictures and/or depth sensors, also by means of machine learning techniques, and finally the process of rendering real or virtual 3D models to standard images and 3D/AR devices.
Students will experience computer vision, deep learning and augmented reality techniques during lab sessions.
A beginners' tutorial on Unity will be provided as well.
Course details will be available soon
The computer vision courses presents the principles and techniques for image processing, understanding and analysis.
The course will show how to extract relevant information from visual data that can be used in challenging real world applications like autonomus driving or smart manifacturing.
It presents the mathematical, programming, and technical issues of these tasks and will include a relevant hands-on laboratory part where students will also develop C++ applications based on the OpenCV library.
e-Health deals with applications of ICT for healthcare and the course covers technologies for acquisition, transmission, and processing of data related to healthcare.
Particularly, it is divided into three units:
(1) data acqusition and analysis, e.g., for biosignals such as EEG and ECG,
(2) communication networks for e-health, e.g., body area sensor networks, and
(3) e-health applications, such as closed-loop interventions and pervasive continuous monitoring.
Hands-on laboratories will be proposed to experience signal acquisition, processing and body area network modelling, as well as guest lectures to meet technical experts and interact also with other professionals involved in the e-health field.
Game theory is the science of analyzing multi-objective multi-agent problems (i.e., "games").
This involves the games we usually play for fun in our everyday life, but in a more serious context is applied to resource competition, distributed management, efficient allocation over multi-user systems and/or communication networks.
This course teaches all the basic concepts, as well as some advanced ones, of game theory.
Also, it applies them to scenarios of interest for ICT.
The course reviews statistical methods applied to large clinical datasets.
The main application is the evaluation of public hygiene data with statistical criteria to assess aspects such as the presence of epidemics, the effectiveness of a therapy, and possible directions and evolutions for treatments.
During the course, this kind of analysis will be also performed on real data taken from statistical datasets in different societal contexts, through laboratory sessions and group assignments.
The course covers the theory and practice of modern artificial neural networks, highlighting their relevance both for machine learning applications and for modeling human cognition and brain function.
Topics include single-neuron modeling and principles of neural encoding; supervised, unsupervised and reinforcement learning; feed-forward and recurrent networks; energy-based models; large-scale brain organization.
Theoretical discussion of various types of network architectures and learning algorithms is complemented by hands-on practices in the computer lab (PyTorch framework).
This is a theoretical course intended to provide knowledge of the main mathematical tools and modeling techniques for the study of telecommunication networks and networking protocols.
The students will get to know the theoretical basics of Markov chains, renewal processes, queueing theory and traffic models.
These instruments will be further applied to the analysis of datalink and networking protocols.
Select 3 courses from the following list
This course adopts a mathematical approach to theory and algorithms for optimization in many fields of network and data analysis. Linear optimization will be explored first, with a review of theory and algorithm (Simplex, interior point).
This is followed by convex optimization, both unconstrained (gradient-type) and constrained (polyhedral approximation, gradient projection).
Finally, the course will discuss large-scale network optimization and clustering methods.
The course targets the principles of user-centered design, cognitive ergonomics, user experience, and usability to investigate how the human experience of interacting with automated computing machines can be made simple, pleasant, and overall satisfactory.
Case studies from websites, apps, smart city applications will be presented and paradigm and design criteria will be reviewed and discussed.
Also, the program will touch accessibility and universal design of interfaces as well as social computing and ergonomics.
The course has three main goals: (i) learning of the techniques and methodologies at the basis of NLP; (ii) development of hard skills for the design of end-to-end NLP systems; (iii) development of soft skills needed for team-working and problem solving.
The course provides knowledge about analytical and synthetic engineering methodologies for the study of the central nervous system.
Covered topics include cerebral hemodynamics (MRI, arterial spin labeling), brain activation maps and connectivity, map generation from PET images, clustering, PCA and ICA, diffusion tensor MRI.
Students will understand potential and limitations of neuroimaging techniques in the study of pathophysiological brain processes.
The integration of physics, statistics, information theory, distributed systems, biochemistry, genetics, and medicine leads to a new research domain with the ambitious goal of giving a physical characterization of organs and living beings.
From an initial focus on the basic molecules of life (DNA, proteins) the course moves to cells, tissues, organs, organisms, and entire ecosystems.
With methodologies borrowed from statistical physics, it explores the complex biochemical processes that are the constituent of life.
The course will offer a carousel of data science methods for the validation of biomarkers in the context of neuroscience studies and for the investigation of brain disorders across the spectrum of neurological and mental health conditions. Specifically, the course will present how biomarkers are used for the development and characterisation of new drugs targeting the central nervous system (including pharmacodynamics and pharmacokinetics properties) as well as for precision medicine applications.
A student successfully completing the course should be able to lay down the key aspects differentiating Reinforcement Learning from other machine learning approaches.
Given an application, the student should be able to determine if it can be adequately formulated as a Reinforcement Learning problem, to be able to formalize it as such and to identify a set of techniques best suited to solve the task, together with the software tools to implement the solution.
This course expands on programming skills already acquired by the students to give a special emphasis on scientific programming.
The students will be guided through the object-oriented programming paradigm to design and develop software in the Python language.
It will be given the competence to analyze formal correctness, computability, and complexity of a program, with a clear problem-solving purpose.
Course details will be available soon
The lab aims to help students improve their oral communication through the study and practice of the elements contributing to successful communication.
The focus is on raising the students' awareness on the importance of verbal and non verbal language in interactions to make communication more effective.
The students will learn the meanings of body language and paralanguage (voice intonation, volume, etc), how they are used in different types of interactions (one-to-one, one-to-many, computer-mediated, etc.), and will have to apply them in a number of assigned tasks.
The lab requires the students' active participation in all class activities, aimed at applying the communication strategies learned.
This course will provide the foundations of the project management.
Traditional (such as the Project Management Institute approach) as well as more advanced techniques - such as the Agile Methodology - will be reviewed.
Special focus will be put on the methodologies more suited for the ICT environment.
Select further 18 ECTS credits from courses of this or another curriculum, or any course of the University of Padova coherently with your overall study plan.
You can select exams from the ICT subjects, the related ones or any other course from the University of Padova coherent with the study plan.
Students are asked to carry out a substantial individual project in their final year.
The project can be carried out either at the University of Padova (30 ECTS combining a 21 ECTS Final Project and a 9 ECTS Report), or in an external institution, such as an Industry or a Research Center, either national or international (30 ECTS combining a 21 ECTS Final Project and a 9 ECTS Internship).
It is also possible to do the internship in an external institution, and the final project at the University, though we suggest to carry out the whole work in a single place.