A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task.

A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task.

Publication date: Mar 07, 2019

Recent studies have demonstrated the use of convolutional neural networks (CNNs) to classify images of melanoma with accuracies comparable to those achieved by board-certified dermatologists. However, the performance of a CNN exclusively trained with dermoscopic images in a clinical image classification task in direct competition with a large number of dermatologists has not been measured to date. This study compares the performance of a convolutional neuronal network trained with dermoscopic images exclusively for identifying melanoma in clinical photographs with the manual grading of the same images by dermatologists.

We compared automatic digital melanoma classification with the performance of 145 dermatologists of 12 German university hospitals. We used methods from enhanced deep learning to train a CNN with 12,378 open-source dermoscopic images. We used 100 clinical images to compare the performance of the CNN to that of the dermatologists. Dermatologists were compared with the deep neural network in terms of sensitivity, specificity and receiver operating characteristics.

The mean sensitivity and specificity achieved by the dermatologists with clinical images was 89.4% (range: 55.0%-100%) and 64.4% (range: 22.5%-92.5%). At the same sensitivity, the CNN exhibited a mean specificity of 68.2% (range 47.5%-86.25%). Among the dermatologists, the attendings showed the highest mean sensitivity of 92.8% at a mean specificity of 57.7%. With the same high sensitivity of 92.8%, the CNN had a mean specificity of 61.1%.

For the first time, dermatologist-level image classification was achieved on a clinical image classification task without training on clinical images. The CNN had a smaller variance of results indicating a higher robustness of computer vision compared with human assessment for dermatologic image classification tasks.

Brinker, T.J., Hekler, A., Enk, A.H., Klode, J., Hauschild, A., Berking, C., Schilling, B., Haferkamp, S., Schadendorf, D., Fr”ohling, S., Utikal, J.S., von Kalle, C., and , Collaborators. A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task. 21923. 2019 Eur J Cancer (111):

Concepts Keywords
Attendings Artificial intelligence
Board Certified Neural network
CNN Sensitivity and specificity
Computer Vision Dermatology
Convolutional Neural Network Deep learning
Convolutional Neural Networks Convolutional neural network
Deep Learning Computer vision
Deep Neural Network Artificial intelligence
Dermatologist Melanoma
Digital Fields of mathematics
Image Classification Articles
Melanoma Academic disciplines
Neuronal Network Computational neuroscience
Open Source Artificial neural networks
Robustness Images melanoma
Variance Convolutional neuronal network
Convolutional neural network

Semantics

Type Source Name
disease DOID Skin cancer
disease MESH Skin cancer
gene UNIPROT LARGE1
gene UNIPROT KCNK3
pathway BSID Melanoma
disease DOID melanoma
disease MESH melanoma
gene UNIPROT JTB
gene UNIPROT NR1I2

Similar

Original Article

Leave a Comment

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