System for neural network recognition of malignant pigmented skin neoplasms with image pre-processing

The article presents a system for the recognition of malignant pigmented skin neoplasms with a preliminary processing stage. Image pre-processing consists of removing hair structures from images, as well as resizing images and their further augmentation. Augmentation made it possible to increase the...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series Vol. 2052; no. 1; pp. 12023 - 12031
Main Authors: Lyakhova, U A, Lyakhov, P A, Abdulkadirov, R I, Efimenko, G A, Romanov, S A, Kaplun, D I
Format: Journal Article
Language:English
Published: IOP Publishing 01.11.2021
ISSN:1742-6588, 1742-6596
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The article presents a system for the recognition of malignant pigmented skin neoplasms with a preliminary processing stage. Image pre-processing consists of removing hair structures from images, as well as resizing images and their further augmentation. Augmentation made it possible to increase the variety of training data, balance the number of images in different categories, and avoid retraining the neural network. The modeling was carried out using the MatLab R2020b software package for solving technical calculations on clinical dermatoscopic images from the international open archive ISIC Melanoma Project. The proposed system for the recognition of malignant pigmented skin neoplasms made it possible to increase the accuracy of image classification up to 80.55%. The use of the proposed recognition system will make it possible to increase the efficiency and quality of diagnosis, in comparison with the methods of visual diagnosis.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/2052/1/012023