Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function

Skin cancer is currently one of the most common types of human cancer. Due to similar morphological manifestations, the diagnosis of malignant neoplasms is difficult even for experienced dermatologists. Artificial intelligence technologies can equal and even surpass the capabilities of an oncologist...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Lyakhova, U. A., Ляхова, У. А.
Formato:	Статья
Lenguaje:	English
Publicado:	2023
Materias:	Cancer Skin lesion analysis Convolutional neural networks Cost-sensitive learning Imbalanced classification Melanoma
Acceso en línea:	https://dspace.ncfu.ru/handle/20.500.12258/25194
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	ir-20.500.12258-25194
record_format	dspace
spelling	ir-20.500.12258-251942023-09-07T08:39:52Z Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function Lyakhova, U. A. Ляхова, У. А. Cancer Skin lesion analysis Convolutional neural networks Cost-sensitive learning Imbalanced classification Melanoma Skin cancer is currently one of the most common types of human cancer. Due to similar morphological manifestations, the diagnosis of malignant neoplasms is difficult even for experienced dermatologists. Artificial intelligence technologies can equal and even surpass the capabilities of an oncologist in terms of the accuracy of visual diagnostics. The available databases of dermoscopic images and statistical data are highly unbalanced about “benign” cases. When training neural network algorithms on unbalanced bases, there is a problem of reducing the accuracy and performance of models due to the prevailing “benign” cases in the samples. One of the possible ways to solve the problem of unbalanced learning is to modify the loss function by introducing different weight coefficients for the recognition classes. The article proposes a neural network system for the recognition of malignant pigmented skin neoplasms, trained using a modified cross-entropy loss function. The accuracy of recognition of malignant neoplasms of the skin in the proposed system was 88.12%. The use of the proposed system by dermatologists-oncologists as an auxiliary diagnostic method will expand the possibilities of early detection of skin cancer and minimize the influence of the human factor. 2023-09-07T08:38:54Z 2023-09-07T08:38:54Z 2023 Статья Lyakhova, U.A. Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function // Lecture Notes in Networks and Systems. - 2023. - 702 LNNS, pp. 353-363. - DOI: 10.1007/978-3-031-34127-4_34 http://hdl.handle.net/20.500.12258/25194 en Lecture Notes in Networks and Systems application/pdf
institution	СКФУ
collection	Репозиторий
language	English
topic	Cancer Skin lesion analysis Convolutional neural networks Cost-sensitive learning Imbalanced classification Melanoma
spellingShingle	Cancer Skin lesion analysis Convolutional neural networks Cost-sensitive learning Imbalanced classification Melanoma Lyakhova, U. A. Ляхова, У. А. Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function
description	Skin cancer is currently one of the most common types of human cancer. Due to similar morphological manifestations, the diagnosis of malignant neoplasms is difficult even for experienced dermatologists. Artificial intelligence technologies can equal and even surpass the capabilities of an oncologist in terms of the accuracy of visual diagnostics. The available databases of dermoscopic images and statistical data are highly unbalanced about “benign” cases. When training neural network algorithms on unbalanced bases, there is a problem of reducing the accuracy and performance of models due to the prevailing “benign” cases in the samples. One of the possible ways to solve the problem of unbalanced learning is to modify the loss function by introducing different weight coefficients for the recognition classes. The article proposes a neural network system for the recognition of malignant pigmented skin neoplasms, trained using a modified cross-entropy loss function. The accuracy of recognition of malignant neoplasms of the skin in the proposed system was 88.12%. The use of the proposed system by dermatologists-oncologists as an auxiliary diagnostic method will expand the possibilities of early detection of skin cancer and minimize the influence of the human factor.
format	Статья
author	Lyakhova, U. A. Ляхова, У. А.
author_facet	Lyakhova, U. A. Ляхова, У. А.
author_sort	Lyakhova, U. A.
title	Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function
title_short	Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function
title_full	Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function
title_fullStr	Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function
title_full_unstemmed	Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function
title_sort	neural network skin cancer recognition with a modified cross-entropy loss function
publishDate	2023
url	https://dspace.ncfu.ru/handle/20.500.12258/25194
work_keys_str_mv	AT lyakhovaua neuralnetworkskincancerrecognitionwithamodifiedcrossentropylossfunction AT lâhovaua neuralnetworkskincancerrecognitionwithamodifiedcrossentropylossfunction
_version_	1809808288074694656

Neural Network Skin Cancer Recognition with a Modified Cross-Entropy Loss Function

Ejemplares similares