Finding reduced Raman spectroscopy fingerprint of skin samples for melanoma diagnosis through machine learning.

Autor: Araújo DC; Computer Science Dept, Federal University of Minas Gerais, Brazil; Kunumi, Brazil. Electronic address: araujodc@ufmg.br., Veloso AA; Computer Science Dept, Federal University of Minas Gerais, Brazil., de Oliveira Filho RS; Dept of Surgery, Division of Plastic Surgery, Federal University of Sao Paulo, Brazil., Giraud MN; Dept of Medicine, University of Fribourg, Switzerland., Raniero LJ; Research and Development Dept, Vale do Para'iba University, Brazil., Ferreira LM; Dept of Surgery, Division of Plastic Surgery, Federal University of Sao Paulo, Brazil., Bitar RA; DERMiSCAN, Brazil. Electronic address: renata.bitar@dermiscan.ch.
Jazyk: angličtina
Zdroj: Artificial intelligence in medicine [Artif Intell Med] 2021 Oct; Vol. 120, pp. 102161. Date of Electronic Publication: 2021 Aug 28.
DOI: 10.1016/j.artmed.2021.102161
Abstrakt: Early-stage detection of cutaneous melanoma can vastly increase the chances of cure. Excision biopsy followed by histological examination is considered the gold standard for diagnosing the disease, but requires long high-cost processing time, and may be biased, as it involves qualitative assessment by a professional. In this paper, we present a new machine learning approach using raw data for skin Raman spectra as input. The approach is highly efficient for classifying benign versus malignant skin lesions (AUC 0.98, 95% CI 0.97-0.99). Furthermore, we present a high-performance model (AUC 0.97, 95% CI 0.95-0.98) using a miniaturized spectral range (896-1039 cm -1 ), thus demonstrating that only a single fragment of the biological fingerprint Raman region is needed for producing an accurate diagnosis. These findings could favor the future development of a cheaper and dedicated Raman spectrometer for fast and accurate cancer diagnosis.
(Copyright © 2021 Elsevier B.V. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: