AI-based radiomics for glioma diagnosis
Reference persons SANTA DI CATALDO
External reference persons email@example.com
Research Groups DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA
Description The early and accurate diagnosis of post-treatment tumor progression (TP) in high-grade gliomas (HGGs) is crucial to improve overall survival (OS) as it can drive optimal therapy. However, new enhanced lesions often appear on MRI in the treatment area which can be due to pseudoprogression (PsP) and treatment-related changes (TRCs).
The aim of the thesis is to develop a novel deep learning-based radiomics tool for the classification of TP, PsP and TRCs leveraging MRI sequences.
The thesis activity, done in strict collaboration with the University of Brescia, will require the following steps:
Literature analysis, looking for the state-of-the-art systems currently available.
Delineation of a preliminary dataset on-line available.
Design of a deep learning-based architecture for the above-mentioned classification task
Optional. Testing/fine-tuning plus testing on a novel cohort of patients, gathered in the mean-while within the collaboration of several hospitals.
Required skills Good programming skills (Python). Biomedical background and prior knowledge of Machine Learning/Deep Learning design frameworks is a plus.
Deadline 28/07/2024 PROPONI LA TUA CANDIDATURA