Risk Segmentation
Risk segmentation research focuses on identifying and categorizing risks, particularly within AI systems and medical image analysis. Current efforts involve developing comprehensive risk taxonomies based on regulatory frameworks and establishing standardized benchmarks for evaluating AI safety, alongside algorithmic advancements in model architectures like U-Net and GANs to improve the accuracy and robustness of automated segmentation in medical imaging. These advancements are crucial for improving AI safety and reliability, as well as enhancing the efficiency and accuracy of medical procedures like radiation therapy planning.
Papers
Ensemble Methods for Multi-Organ Segmentation in CT Series
Leonardo Crespi, Paolo Roncaglioni, Damiano Dei, Ciro Franzese, Nicola Lambri, Daniele Loiacono, Pietro Mancosu, Marta Scorsetti
Comparing Adversarial and Supervised Learning for Organs at Risk Segmentation in CT images
Leonardo Crespi, Mattia Portanti, Daniele Loiacono