Please use this identifier to cite or link to this item:
http://hdl.handle.net/10553/121432
Title: | Attention-based Skin Cancer Classification Through Hyperspectral Imaging | Authors: | La Salvia, Marco Torti, Emanuele Gazzoni, Marco Marenzi, Elisa León Martín, Sonia Raquel Ortega Sarmiento,Samuel Fabelo Gómez, Himar Antonio Marrero Callicó, Gustavo Iván Leporati, Francesco |
UNESCO Clasification: | Investigación | Keywords: | Deep learning Medical hyperspectral imaging Skin cancer Vision Transformers |
Issue Date: | 2022 | Publisher: | Institute of Electrical and Electronics Engineers Inc. | Conference: | 25th Euromicro Conference on Digital System Design (DSD 2022) | Abstract: | In recent years, hyperspectral imaging has been employed in several medical applications, targeting automatic diagnosis of different diseases. These images showed good performance in identifying different types of cancers. Among the methods used for classification, machine learning and deep learning techniques emerged as the most suitable algorithms to handle these data. In this paper, we propose a novel hyperspectral image classification architecture exploiting Vision Transformers. We validated the method on a real hyperspectral dataset containing 76 skin cancer images. Obtained results clearly highlight that the Vision Transforms are a suitable architecture for this task. Measured results outperform the state-of-the-art both in terms of false negative rates and of processing times. Finally, the attention mechanism is evaluated for the first time on medical hyperspectral images. | URI: | http://hdl.handle.net/10553/121432 | ISBN: | 9781665474047 | DOI: | 10.1109/DSD57027.2022.00122 | Source: | Proceedings - 2022 25th Euromicro Conference on Digital System Design, DSD 2022 / Himar Fabelo, Samuel Ortega, Amund Skavhaug (Eds.), p. 871-876 |
Appears in Collections: | Actas de congresos |
SCOPUSTM
Citations
7
checked on Mar 30, 2025
WEB OF SCIENCETM
Citations
5
checked on Mar 30, 2025
Google ScholarTM
Check
Altmetric
Share
Export metadata
Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.