Melanoma recognition by a deep learning convolutional neural network-Performance in different melanoma subtypes and localisations.

Publication date: Jan 20, 2020

Deep learning convolutional neural networks (CNNs) show great potential for melanoma diagnosis. Melanoma thickness at diagnosis among others depends on melanoma localisation and subtype (e.g. advanced thickness in acrolentiginous or nodular melanomas). The question whether CNN may counterbalance physicians’ diagnostic difficulties in these melanomas has not been addressed. We aimed to investigate the diagnostic performance of a CNN with approval for the European market across different melanoma localisations and subtypes.

The current market version of a CNN (Moleanalyzer-Pro(R), FotoFinder Systems GmbH, Bad Birnbach, Germany) was used for classifications (malignant/benign) in six dermoscopic image sets. Each set included 30 melanomas and 100 benign lesions of related localisations and morphology (set-SSM: superficial spreading melanomas and macular nevi; set-LMM: lentigo maligna melanomas and facial solar lentigines/seborrhoeic keratoses/nevi; set-NM: nodular melanomas and papillomatous/dermal/blue nevi; set-Mucosa: mucosal melanomas and mucosal melanoses/macules/nevi; set-AM: acrolentiginous melanomas and acral (congenital) nevi; set-AM: subungual melanomas and subungual (congenital) nevi/lentigines/ethnical type pigmentations).

The CNN showed a high-level performance in set-SSM, set-NM and set-LMM (sensitivities >93.3%, specificities >65%, receiver operating characteristics-area under the curve [ROC-AUC] >0.926). In set-AM, the sensitivity was lower (83.3%) at a high specificity (91.0%) and ROC-AUC (0.928). A limited performance was found in set-mucosa (sensitivity 93.3%, specificity 38.0%, ROC-AUC 0.754) and set-AM (sensitivity 53.3%, specificity 68.0%, ROC-AUC 0.621).

The CNN may help to partly counterbalance reduced human accuracies. However, physicians need to be aware of the CNN’s limited diagnostic performance in mucosal and subungual lesions. Improvements may be expected from additional training images of mucosal and subungual sites.

Winkler, J.K., Sies, K., Fink, C., Toberer, F., Enk, A., Deinlein, T., Hofmann-Wellenhof, R., Thomas, L., Lallas, A., Blum, A., Stolz, W., Abassi, Fuchs, T., Rosenberger, A., and Haenssle, H.A. Melanoma recognition by a deep learning convolutional neural network-Performance in different melanoma subtypes and localisations. 25541. 2020 Eur J Cancer (127):

Concepts Keywords
AM Convolutional neural network
AUC Melanoma
Benign Organ systems
CNN Dermatology
Congenital Integumentary system
Convolutional Neural Network Nevus
Convolutional Neural Networks Lentigo maligna melanoma
Deep Learning Melanocytic nevus
Germany Dermatoscopy
Lentigines Congenital melanocytic nevus
Level Set Mucosal melanoma
Macules Nodular melanoma
Malignant Neural network
Melanoma
Melanomas
Morphology
Mucosa
Nevi
ROC

Semantics

Type Source Name
disease MESH Melanoma
pathway KEGG Melanoma
disease MESH diagnosis
drug DRUGBANK Methyprylon
disease MESH nevi
disease MESH lentigo maligna
disease MESH lentigines
disease MESH keratoses
disease MESH blue nevi
disease MESH melanoses
disease MESH congenital
drug DRUGBANK Saquinavir

Similar

Original Article

Leave a Comment

Your email address will not be published. Required fields are marked *