Il presente modulo ha come obiettivo quello di fornire allo studente gli elementi fondamentali relativi all'elaborazione di immagini mediche. L'imaging medicale si caratterizza, nell'ambito piu ampio dell'image processing, per la necessità di una maggiore attenzione agli aspetti più clinici dell'immagine, primo fra tutti la valenza diagnostica dell'immagine medesima. Il medico, infatti, rimane il fruitore ultimo dell'immagine medica, che si configura come uno strumento con piu valenze: diagnostica, terapeutica, di monitoraggio, di ricerca e di validazione. Non si intende confinare il corso alle immagini di diagnostica umana in-vivo, ma si intende comprendere anche le applicazioni piu collegate alle scienze della vita, quali l?elaborazione delle immagini cellulari e sub-cellulari, precliniche e molecolari. Scopo del corso, quindi, e quello di affrontare le problematiche tipiche connesse all'utilizzo di immagini nelle diverse applicazioni delle scienze della vita. Obiettivo primario e quello di fornire allo studente una "visione" dell'elaborazione che vada al di là delle tecniche numeriche e computazionali, ma che tenga il quesito clinico (o scientifico) sempre come target primario. Obiettivi correlati sono la capacita di individuare le tecniche di elaborazione piu adatte nelle diverse condizioni.

Knowledge and Understanding In the specific field of medical imaging, foundational knowledge includes the ability to understand the nature and diagnostic value of an image, which always represents only a partial view of a physical/physiological reality. Applying Knowledge and Understanding By the end of the course, students should be able to: • Distinguish different types of medical images, • Match processing techniques with diagnostic/scientific questions, • Develop innovative strategies and solutions accordingly. Judgment Autonomy This course helps develop independent judgment skills, particularly during laboratory sessions. Communication Skills The course also aims to improve students' written and oral communication skills through lab activities. Learning Skills The course encourages continuous learning by encouraging students to consult and read scientific articles published in international journals. This activity aims to provide knowledge of the most current topics in the field of medical image processing. Moreover, the lab sessions are fundamental for putting into practice the theoretical concepts.

Students are expected to already possess basic knowledge of digital data processing. In particular, knowledge of digital filter theory is essential. A solid foundation in geometry is important for optimal understanding of some of the techniques presented. Students should also be familiar with basic physiology and anatomy, as well as the physical principles behind common medical imaging devices.

The goal is to equip students with the basic knowledge required to process clinically relevant images for diagnostic or computational purposes. Students will learn how to enhance the diagnostic value of an image by: • Reducing noise, • Modifying edge appearance, • Equalizing color scales, • Improving detail perception. From a computational perspective, students will learn how to: • Segment images, • Calculate areas and perimeters, • Reconstruct 3D volumes. The course will also introduce common tomographic visualization and computer vision techniques. A broad range of clinical applications will be presented, covering morphological and functional images, across scales ranging from cells to whole organs or systems. Main Course Topics: • Images, the eye, and our visual system. Shape and color perception. • Key parameters of a clinical image and their diagnostic relevance: brightness, optical density, contrast, spatial resolution, depth resolution, amplitude resolution. • Point-wise and local operations. Brightness and contrast adjustment. Automatic image equalization. Gamma correction. • 2D filters. Low-pass and high-pass filters. Edge detection. Median filters. Specific denoising techniques. • Introduction to image segmentation. • Elements of artificial intelligence applied to medical imaging: deep learning architectures for segmentation. • Other image segmentation techniques: Morphological techniques. Histogram-based methods. Parametric and geometric deformable models. • ITK/VTK and professional solutions for processing and re-visualization applications. • 3D perfusion and vascularization studies. Vascular network skeletonization. • Initial concepts on designing and developing a software system that processes medical images.

The course includes lectures and computer-based lab sessions. Labs aim to provide students with real images to which they can apply the techniques covered in class. Lab work will typically be done in groups of four. The programming environments used will be Python and C/C++ (required for specific libraries). A voluntary contest will be held during the course to optimize a processing code. Each group will be provided with a shared dataset and tasked with developing a solution. The performance of each group’s algorithm will be evaluated to understand strengths and weaknesses of different approaches.

• Slides provided by the instructor • Recently published scientific articles on relevant techniques

Slides; Video lezioni tratte da anni precedenti; Materiale multimediale ;

Modalità di esame: Prova orale obbligatoria; Elaborato progettuale in gruppo;

Exam: Compulsory oral exam; Group project;

... Onsite Exam: • Mandatory oral exam • Group project Each group (4 students) is assigned a dataset and a specific task, typically aligned with current challenges in medical imaging. Groups must submit: 1. Code solving the assigned task 2. A report summarizing the methodology and results Grading: • Project solution: up to 12 points • Report quality: up to 10 points • Oral exam (individual): up to 10 points Final grade = sum of the three components. Honors (lode) is awarded if total > 30 points.

Gli studenti e le studentesse con disabilità o con Disturbi Specifici di Apprendimento (DSA), oltre alla segnalazione tramite procedura informatizzata, sono invitati a comunicare anche direttamente al/la docente titolare dell'insegnamento, con un preavviso non inferiore ad una settimana dall'avvio della sessione d'esame, gli strumenti compensativi concordati con l'Unità Special Needs, al fine di permettere al/la docente la declinazione più idonea in riferimento alla specifica tipologia di esame.