no code implementations • 11 Jan 2022 • Andrea Leo, Giacomo Reatti, Stefano Arrigoni, Michael Khayyat, Federico Cheli
It is estimated that 90% of crashes occur due to human error, mainly induced by poor judgement, distraction or lack of situation awareness.
2 code implementations • 26 Aug 2021 • Gabriele Valvano, Andrea Leo, Sotirios A. Tsaftaris
After training is complete, the discriminator is usually discarded, and only the generator is used for inference.
1 code implementation • 26 Aug 2021 • Gabriele Valvano, Andrea Leo, Sotirios A. Tsaftaris
Collecting large-scale medical datasets with fine-grained annotations is time-consuming and requires experts.
2 code implementations • 26 Aug 2021 • Gabriele Valvano, Andrea Leo, Sotirios A. Tsaftaris
At inference, the discriminator is discarded, and only the segmentor is used to predict label maps on test images.
3 code implementations • 2 Jul 2020 • Gabriele Valvano, Andrea Leo, Sotirios A. Tsaftaris
We evaluated our model on several medical (ACDC, LVSC, CHAOS) and non-medical (PPSS) datasets, and we report performance levels matching those achieved by models trained with fully annotated segmentation masks.
1 code implementation • 29 Aug 2019 • Gabriele Valvano, Agisilaos Chartsias, Andrea Leo, Sotirios A. Tsaftaris
There has been an increasing focus in learning interpretable feature representations, particularly in applications such as medical image analysis that require explainability, whilst relying less on annotated data (since annotations can be tedious and costly).
no code implementations • 29 Oct 2018 • Gabriele Valvano, Andrea Leo, Daniele Della Latta, Nicola Martini, Gianmarco Santini, Dante Chiappino, Emiliano Ricciardi
Recent research put a big effort in the development of deep learning architectures and optimizers obtaining impressive results in areas ranging from vision to language processing.
1 code implementation • 25 Oct 2018 • Gabriele Valvano, Nicola Martini, Andrea Leo, Gianmarco Santini, Daniele Della Latta, Emiliano Ricciardi, Dante Chiappino
Skull-stripping methods aim to remove the non-brain tissue from acquisition of brain scans in magnetic resonance (MR) imaging.
no code implementations • 2 Jul 2018 • Gianmarco Santini, Lorena M. Zumbo, Nicola Martini, Gabriele Valvano, Andrea Leo, Andrea Ripoli, Francesco Avogliero, Dante Chiappino, Daniele Della Latta
In Europe the 20% of the CT scans cover the thoracic region.