Course Structure

The MAIA Master is a two-year programme of study divided into four semesters designed and developed to ensure a clear and structured educational progress in the field of medical image analysis, fully integrated among the three universities of the consortium:

A progressive mobility pathway

MAIA has been designed to ensure a clear, coherent, and progressive educational pathway, fully integrated across the three consortium universities. Students follow a mandatory cohort-based mobility scheme, moving together from one institution to the next each semester.

What you study

Year 1

Semester 1 – Université Bourgogne Europe

Students acquire solid foundations in signal and image processing, machine learning, and medical imaging.

Semester 2 – Università degli Studi di Cassino e del Lazio Meridionale

Students deepen their knowledge in advanced computer vision and machine and deep learning methods.

Topics/Contents • Physical principles of medical imaging • Image formation and acquisition • Magnetic resonance imaging (MRI) • Computed tomography (CT) • Positron emission tomography (PET) • Ultrasound imaging • Image quality and quality assessment

Topics / Contents • Probability and statistics • Optimisation methods • Regularisation • Information theory • Mathematical foundations of machine learning and medical AI

Topics / Contents • Fundamentals of robotics for medical AI • Coordinate systems • Rigid transformations • Homographies • Sensor models • Robot perception

Topics / Contents • Image filtering • Image enhancement • Image segmentation • Image registration • Feature extraction • Classical medical image processing

Topics / Contents • Python programming • Scientific Python libraries • Medical image I/O and manipulation • Data analysis and visualisation • Reproducible computational workflows • Programming practices

Topics / Contents • Critical reading of scientific literature • Methodological analysis • Evidence assessment • Scientific argumentation • Discussion and presentation of research findings

Topics / Contents • Regression • Classification • Ensemble methods • Model selection • Model evaluation • Machine learning for medical datasets

Topics / Contents • Deep learning architectures • Training and optimisation • Medical image analysis • Object detection • Deep learning for image registration • Model evaluation

Topics / Contents • Model-based and data-driven robotics • Visual servoing • Human–robot interaction • Safety-critical control • Automatic control and machine learning • Adaptive medical robots

Topics / Contents • Classical segmentation • Learning-based segmentation • Segmentation uncertainty • Segmentation evaluation • Clinical relevance of segmentation outputs

Topics / Contents • EU AI Act • Medical Device Regulation (MDR) • FDA regulatory pathways • Explainability and interpretability • Fairness and bias • Accountability • Risk management • Trustworthy AI

Topics / Contents • Innovation processes • Technology transfer • Translation of medical AI • Multidisciplinary teamwork • Project pitching • Stakeholder communication

Year 2

Semester 3 – Universitat de Girona

The programme focuses on applied, hands-on modules in computer-aided diagnosis and medical applications.

Semester 4 – Master Thesis Host Institution

Students carry out their Master’s thesis at one of the consortium universities or at an associated partner institution, including research centres, universities or companies.

Topics / Contents • State-of-the-art AI • Generative models • Diffusion models • Emerging architectures • Advanced medical imaging applications • Current research challenges

Topics / Contents • Multimodal medical data • Integration of imaging and clinical variables • Longitudinal studies • Temporal analysis • Disease progression modelling

Topics / Contents • Self-supervised learning • Contrastive learning • Masked autoencoders • Transfer learning • Foundation models • Model adaptation • Vision–language models

Topics / Contents • Image-guided interventions • Intraoperative imaging • Video analysis • Pre-operative registration • Intra-operative registration • AI-based decision support

Topics / Contents • Evaluation methodologies • External validation • Reproducibility • Benchmarking • Reporting standards • Software as a Medical Device (SaMD) • Clinical evidence • Regulatory documentation

Topics / Contents • Regulatory challenges • Professional responsibilities • Deployment of AI in healthcare • Case-based discussions • Decision-making scenarios

Topics / Contents • Problem definition • Literature review and research methodology • Data analysis • Methodological development • Experimental validation • Scientific writing • Oral defence

Topics / Contents • Scientific writing • Structure of research papers • Journal and conference selection • Peer review • Responding to reviewers • Publication of thesis-related research

Topics / Contents • Cloud, edge, and mobile deployment • MLOps fundamentals • Cybersecurity risks • Threat modelling • Data protection • GDPR in healthcare AI

Topics / Contents • Academic careers • Industrial and MedTech careers • PhD preparation • Career planning • Funding opportunities • Innovation and technology transfer • Professional profile

[ Download the full programme (PDF) ]

Joint academic activities

Induction Week · September · Bourgogne

An induction week is organised each year in September, during which enrolled students are introduced to the programme, assessment procedures and academic rules. The induction week also provides an opportunity for new students to meet MAIA graduate students.

Winter School · February · Cassino

In addition to these modules, a Winter School is organised every February in Cassino, serving as a key joint academic milestone of MAIA. The Winter School brings together students from both the first and second year across the consortium, with differentiated participation modes reflecting their respective stages of academic progression.

First-year students attend on site, while second-year students participate online, as they are already engaged in their Master’s thesis activities.

Beyond its teaching function, the Winter School acts as a shared European learning space that strengthens programme integration across partner institutions. It also serves as the formal launch of the online transversal modules of the fourth semester.

MAIA Day · June · Girona

MAIA Day is an annual event celebrating the programme and its international research community. It features presentations of students’ work and provides a meeting point for current students, graduates, researchers and potential collaborators from academia and industry.

Graduation Ceremony

The Graduation Ceremony celebrates the achievements of students completing the MAIA programme. It marks the culmination of their academic journey and recognises the knowledge, skills and international experience gained throughout the master’s degree.

Master's Thesis

The Master’s Thesis allows students to apply the knowledge and methods acquired during the programme to a research or development project in medical image analysis. It is an opportunity to explore a specialised topic in depth and demonstrate their ability to address challenges in medical imaging.

Seminars

MAIA seminars provide opportunities to explore current research and emerging developments in medical image analysis. Through talks by researchers and specialists, students can broaden their knowledge, discover new research directions and engage with topics beyond the regular curriculum.

Programme Framework

The curriculum is structured around three complementary pillars:

01 — Medical image acquisition, analysis and image-guided interaction with the physical world

Medical image acquisition, analysis and image-guided interaction with the physical world.

02 — AI-driven modelling, learning and perception-action techniques

AI-driven modelling, learning and perception-action techniques.

03 — clinical validation, regulatory and ethical aspects of medical image analysis and intervention.

Clinical validation, regulatory and ethical aspects of medical image analysis and intervention.

Progressive Learning

These pillars are addressed in a transversal manner and progressively reinforced across all semesters, with increasing depth, integration, and practical relevance as students advance through the programme.

Programme Outcomes

Together, this structure ensures that graduates acquire both advanced technical expertise and the transversal competences required for the responsible design, validation and deployment of medical AI systems in real-world healthcare settings.

Apply to MAIA

Click the button below to begin the application process

Start your application

OPEN SOON!

Scroll to Top

Learn how we helped