Attention-based neural network for multiclass classification of human skin carcinoma using in vivo autofluorescence and diffuse reflectance spectroscopy

Ni D, Camonin M, Kupriyanov V, Amouroux M, Khairallah G, Blondel W, Hohmann M (2025)

Publication Type: Conference contribution

Publication year: 2025

Journal

Progress in Biomedical Optics and Imaging - Proceedings of SPIE

Publisher: SPIE

Book Volume: 13934

Conference Proceedings Title: Progress in Biomedical Optics and Imaging - Proceedings of SPIE

Event location: Munich, DEU

ISBN: 9781510698055

DOI: 10.1117/12.3097791

Abstract

This contribution presents an attention-based neural network model developed to classify skin carcinomas, actinic keratosis and normal skin from multimodal spectroscopy on 131 patients. Preliminary results show accuracy and f1-score of 64 % and 0.44 respectively.

Authors with CRIS profile

Dongqin Ni Lehrstuhl für Photonische Technologien (LPT) Martin Hohmann Lehrstuhl für Photonische Technologien (LPT)

Involved external institutions

Le Centre Hospitalier Régional de Metz-Thionville FR France (FR) Université de Lorraine FR France (FR)

How to cite

APA:

Ni, D., Camonin, M., Kupriyanov, V., Amouroux, M., Khairallah, G., Blondel, W., & Hohmann, M. (2025). Attention-based neural network for multiclass classification of human skin carcinoma using in vivo autofluorescence and diffuse reflectance spectroscopy. In Zhiwei Huang, Lothar D. Lilge (Eds.), Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Munich, DEU: SPIE.

MLA:

Ni, Dongqin, et al. "Attention-based neural network for multiclass classification of human skin carcinoma using in vivo autofluorescence and diffuse reflectance spectroscopy." Proceedings of the 4th Translational Biophotonics: Diagnostics and Therapeutics, Munich, DEU Ed. Zhiwei Huang, Lothar D. Lilge, SPIE, 2025.

BibTeX: Download