![]() |
JOURNAL TOOLS |
Opzioni di pubblicazione |
eTOC |
Per abbonarsi |
Sottometti un articolo |
Segnala alla tua biblioteca |
ARTICLE TOOLS |
Publication history |
Estratti |
Permessi |
Per citare questo articolo |
Share |


I TUOI DATI
I TUOI ORDINI
CESTINO ACQUISTI
N. prodotti: 0
Totale ordine: € 0,00
COME ORDINARE
I TUOI ABBONAMENTI
I TUOI ARTICOLI
I TUOI EBOOK
COUPON
ACCESSIBILITÀ
REVIEW MELANOMA DIAGNOSIS AND MANAGEMENT Free access
Italian Journal of Dermatology and Venereology 2021 June;156(3):289-99
DOI: 10.23736/S2784-8671.20.06753-X
Copyright © 2020 EDIZIONI MINERVA MEDICA
lingua: Inglese
Artificial intelligence for melanoma diagnosis
Philipp TSCHANDL ✉
Department of Dermatology, Medical University of Vienna, Vienna, Austria
Convolutional neural networks (CNN) have shown unprecedented accuracy in digital image analysis, which can be harnessed for melanoma recognition through automated evaluation of clinical and dermatoscopic images. In experimental studies, modern CNN architectures perform single image analysis at the level of dermatologists and domain-experts, also for multiclass predictions including a multitude of possible diagnoses. This may not necessarily translate to good clinical performance, and reliable randomized controlled prospective clinical trials for modern CNNs are essentially missing. Weaknesses of CNNs are that limitations of available training image datasets propagate to limitations of CNN predictions, and they cannot provide a reliable estimate of uncertainty. Recent research focuses on human-computer collaboration, where gains in accuracy were measured even with imperfect CNNs. With missing academic and clinical agreement on equivocal melanocytic lesions, fully automating histologic assessment of them with CNNs appear problematic, and applications in the near future are probably limited to supporting, referencing or recommendation roles.
KEY WORDS: Melanoma; Diagnosis; Dermoscopy; Artificial intelligence; Neural networks, computer; Machine learning