The effective and non-invasive diagnosis of skin cancer is a hot topic in biophotonics since the current gold standard, a biopsy, is slow and costly. Non-invasive optical techniques such as polarization and multispectral imaging have arisen as powerful tools to overcome these constraints. The combination of these techniques provides a comprehensive characterization of skin chromophores including polarization, color and spectral features. Hence, in this work we propose a polarized multispectral imaging device that works from 414 nm to 995 nm and at 0°, 45° and 90° polarization configurations. Preliminary results performed over 20 nevi and 20 melanoma found statistically significant descriptors (p<0.05) that discriminated between these two lesion etiologies. A further analysis of more lesions is expected to contribute in reducing the false positives during the diagnosis process and, as a consequence, the number of necessary biopsies.
Polarized multispectral imaging for the diagnosis of skin cancer / Rey-Barroso, L.; Burgos-Fernandez, F. J.; Royo, S.; Puig, S.; Malvehy, J.; Pellacani, G.; Delpueyo, X.; Pena, S.; Diaz-Douton, F.; Vilaseca, M.. - 2019-October:(2019), pp. 381-385. (Intervento presentato al convegno 27th Color and Imaging Conference: Color Science and Engineering Systems, Technologies, and Applications, CIC 2019 tenutosi a fra).
Polarized multispectral imaging for the diagnosis of skin cancer
Pellacani G.;
2019
Abstract
The effective and non-invasive diagnosis of skin cancer is a hot topic in biophotonics since the current gold standard, a biopsy, is slow and costly. Non-invasive optical techniques such as polarization and multispectral imaging have arisen as powerful tools to overcome these constraints. The combination of these techniques provides a comprehensive characterization of skin chromophores including polarization, color and spectral features. Hence, in this work we propose a polarized multispectral imaging device that works from 414 nm to 995 nm and at 0°, 45° and 90° polarization configurations. Preliminary results performed over 20 nevi and 20 melanoma found statistically significant descriptors (p<0.05) that discriminated between these two lesion etiologies. A further analysis of more lesions is expected to contribute in reducing the false positives during the diagnosis process and, as a consequence, the number of necessary biopsies.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
