Thematic Session 9

THEMATIC SESSION #09

Giorgio De Nunzio

University of Salento

• giorgio.denunzio@unisalento.it

Rocco Rizzo

University of Salento

• rocco.rizzo1@unisalento.it

Radiomics is rapidly transforming biomedical imaging into a powerful source of quantitative biomarkers, supporting diagnosis, prognosis, and treatment planning in modern healthcare.

This Special Session focuses on the integration of radiomics with Artificial Intelligence, covering both traditional machine learning methods based on handcrafted imaging features and deep learning approaches capable of learning representations directly from medical images. Applications span multiple modalities, including CT, MRI, and PET, enabling data-driven characterization of disease and therapy response.

The session will highlight recent advances across radiomics-based pipelines, deep neural models, and emerging combinations of the two, while also addressing key challenges such as robustness, interpretability, and clinical translation. Overall, the session aims to showcase how AI-driven medical imaging is shaping the future of precision medicine.

In addition, the session welcomes contributions exploring how AI-driven imaging biomarkers can support emerging clinical technologies, including visualization and interaction frameworks enabled by Extended Reality (XR). In such contexts, intelligent imaging models may enhance immersive environments for clinical decision support, education, and patient-specific planning.

Submissions are encouraged on, but not limited to, the following topics:

• Radiomics feature extraction and quantitative imaging biomarkers in clinical practice
• Machine learning methods for radiomics-based diagnosis, prognosis, and risk stratification
• Deep learning architectures for medical image analysis (e.g., CNNs, Transformers)
• End-to-end AI models for automated detection, segmentation, and classification in medical imaging
• Radiomics and AI applications across imaging modalities (CT, MRI, PET, ultrasound)
• Hybrid and integrated approaches combining handcrafted radiomic features with deep neural representations
• Multimodal learning: integration of imaging with clinical, genomic, or electronic health record data
• Explainable AI and interpretability in radiomics and deep learning for healthcare
• Robustness, reproducibility, and standardization of radiomics and AI pipelines
• Clinical validation, translational studies, and real-world deployment of AI-driven imaging biomarkers
• AI-assisted therapy response assessment and outcome prediction
• Emerging trends in precision medicine through intelligent medical imaging
• Intelligent imaging biomarkers for immersive visualization and XR-supported clinical workflows
• Extended Reality environments enhanced by AI-based medical image understanding
• Emerging trends in precision medicine through intelligent medical imaging and interactive healthcare technologies

Giorgio De Nunzio is a researcher and adjunct professor at the Department of Mathematics and Physics "Ennio De Giorgi" at the University of Salento (Lecce, Italy), specializing in Applied Physics since 2001. He earned his degree in Physics in Lecce (1991) and completed a PhD at the University of Montpellier II (1995). His research focuses on the application of Physics and Informatics to Medicine (e.g., diagnostic imaging, Artificial Intelligence) and Cultural Heritage. He leads the Laboratory of Biomedical Physics and Environment at the University of Salento and coordinates the ADAM group (http://adam.unisalento.it). His teaching activities include courses in Physics and AI for Medicine, Machine Learning, C++, and Python.

Rocco Rizzo graduated with a Master's degree in Physics (LM-17) from the University of Salento in 2024. He is a research fellow at the AVR Lab (Augmented and Virtual Reality Laboratory) within the Department of Innovation Engineering at the University of Salento. His research interests include artificial intelligence applied to medicine, with a particular focus on machine learning and deep learning for medical image analysis and diagnostic classification.