Use of Neural Network-Based Deep Learning Techniques for the Diagnostics of Skin Diseases


Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription Access

Abstract

Melanoma is one of the most dangerous types of cancer. The accuracy of visual diagnosis of melanoma directly depends on the experience and specialty of the physician. Current development of image processing and machine learning technologies allows systems based on artificial neural convolutional networks to be created, these being better than humans in object classification tasks, including the diagnostics of malignant skin neoplasms. Presented here is an algorithm for the early diagnostics of melanoma based on artificial deep convolutional neural networks. This algorithm can discriminate benign and malignant skin tumors with an accuracy of at least 91% by examination of dermatoscopy images.

About the authors

D. A. Gavrilov

Moscow Institute of Physics and Technology

Author for correspondence.
Email: gavrilou@gmail.com
Russian Federation, Dolgoprudny, Moscow Region

A. V. Melerzanov

Moscow Institute of Physics and Technology

Email: gavrilou@gmail.com
Russian Federation, Dolgoprudny, Moscow Region

N. N. Shchelkunov

Moscow Institute of Physics and Technology

Email: gavrilou@gmail.com
Russian Federation, Dolgoprudny, Moscow Region

E. I. Zakirov

Moscow Institute of Physics and Technology

Email: gavrilou@gmail.com
Russian Federation, Dolgoprudny, Moscow Region

Supplementary files

Supplementary Files
Action
1. JATS XML

Copyright (c) 2019 Springer Science+Business Media, LLC, part of Springer Nature