Cancer research has significantly advanced in recent yearsmainly through developments in medical genomics andbioinformatics. It is expected that such approaches will resultin more durable tumor control and fewer side effects comparedwith conventional treatments such as radiotherapy orchemotherapy. From the imaging standpoint, non-invasive imagingbiomarkers (IBs) that assess angiogenic response and tumorenvironment at an early stage of therapy are of utmostimportance, since they could provide useful insights into therapyplanning. However, the extraction of IBs is still an openproblem, since there are no standardized imaging protocols yet orestablished methods for the robust extraction of IBs. DCE-MRI isamongst the most promising non-invasive functional imagingmodalities with compartmental pharmacokinetic (PK) modeling beingthe most common technique used for DCE-MRI data analysis.However, PK models suffer from a number of limitations such asmodeling complexity, which often leads to variability in thecomputed biomarkers. To address these problems, alternativeDCE-MRI biomarker extraction strategies coupled with a profoundunderstanding of the physiological meaning of IBs is a sine quanon condition. To this end, a more recent model-free approach hasbeen suggested in literature for DCE-MRI data analysis, whichrelies on the shape classification of the time-signal uptakecurves of image pixels in a selected tumor region of interest.This talk is centered on this classification approach and theclinical question whether model-free DCE-MRI data analysis hasthe potential to provide robust, clinically significantbiomarkers using pattern recognition and image analysistechniques.
Latest Announcements
Seminars
Giovani in un'ora - Ciclo di seminari - Quinta parte
2024-11-27
Giulio Del Corso - "Generating a physically accurate cardiac MRI: a story of (interesting) failures and (justifiable) numerical shortcuts"Abstract: "The generation of physically accurate cardiac-MRI video sequences is a challenging topic that requires a combination of modern g...
Giovani in un'ora - Ciclo di seminari - Quarta parte
2024-11-20
Luca Ciampi - "Mind the Prompt: A Novel Benchmark for Prompt-based Class-Agnostic Counting"Abstract: "Object counting estimates the number of objects in images or video frames. Studies reveal that the human brain employs two distinct methods for counting objects owing to the s...
Giovani in un'ora - Ciclo di seminari - Terza parte
2024-10-16
Gabriele Lagani - "Hebbian learning algorithms for deep neural networks: explorations and outlooks"Abstract: Deep learning systems have achieved outstanding results in various AI tasks. However, such system suffer from a number of limitations, for example in terms of energy an...
Giovani in un'ora - Ciclo di seminari - Seconda parte
2024-10-09
Saira Bano - "From Complexity to Clarity: Enhancing Cross-Modal Knowledge Distillation via Multimodal Teacher Ensembles"Abstract: Traditional knowledge distillation (KD) typically uses a large, complex teacher model, often trained to a single modality, to transfer knowledge to...
Patient Interaction – for well-being, productivity and sustainability
2024-10-08
We live in aworld of instant results and fleeting gratification. In HCI no less: the designprinciples for direct manipulation require immediate feedback and, in the caseof graphical actions, sub-second responses. In addition, computers expect us togive them our undivided atten...
Giovani in un'ora - Ciclo di seminari - Prima parte
2024-10-02
Ali Reza Omrani - "Machine Learning to Measure Vocal Stereotypy: An Extension"Abstract: Repeated measurement of behavior is a process central to behavior analysis, but its implementation occasionally requires hiring observers dedicated exclusively to data collection, which may...
A general framework for distributed approximate similarity search with arbitrary distances
2024-09-26
While many similarity search algorithms are specifically adapted to metric distances,they are unsuitable for alternatives like the cosine distance, which has gained popularity, particularly with embeddings and text mining. To address thisissue, we propose GDASC (General Distri...
In this talk, I will focus on a connection between stable-failures refinement and the ioco conformance relation. Both behavioural relations underlie methodologies that have gained traction in industry: stable-failures refinement is used in several commercial Model-Driven Engin...
Explainability in deep learning models applied to spatio-temporal problems
2024-09-13
Artificial Intelligence (AI) is transforming society, affecting everything from industry to decision making, and concerns about its transparency have increased. Explainable Artificial Intelligence (XAI) is crucial to address this problem, allowing to obtain a better understand...
Making 5G Networks Reliable for Next-generation Applications using AI
2024-05-27
The emergence of 5G technology marks a significant milestone in developing telecommunication networks, enabling exciting new applications such as augmented reality and self-driving vehicles. However, these improvements bring an increased management complexity and a special con...
