Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)

Deep learning has gained immense attention from researchers in medicine, especially in medical imaging. The main bottleneck is the unavailability of sufficiently large medical datasets required for the good performance of deep learning models. This paper proposes a new framework consisting of one va...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Bilal Ahmad, Sun Jun, Vasile Palade, Qi You, Li Mao, Mao Zhongjie
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	variational autoencoder generative adversarial networks melanoma detection skin cancer classification student t-distribution heavy-tailed distribution Medicine (General) R5-920
Acceso en línea:	https://doaj.org/article/3b9db49dbab84d11853d3f1754a2ff87
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:3b9db49dbab84d11853d3f1754a2ff87
record_format	dspace
spelling	oai:doaj.org-article:3b9db49dbab84d11853d3f1754a2ff872021-11-25T17:22:02ZImproving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)10.3390/diagnostics111121472075-4418https://doaj.org/article/3b9db49dbab84d11853d3f1754a2ff872021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4418/11/11/2147https://doaj.org/toc/2075-4418Deep learning has gained immense attention from researchers in medicine, especially in medical imaging. The main bottleneck is the unavailability of sufficiently large medical datasets required for the good performance of deep learning models. This paper proposes a new framework consisting of one variational autoencoder (VAE), two generative adversarial networks, and one auxiliary classifier to artificially generate realistic-looking skin lesion images and improve classification performance. We first train the encoder-decoder network to obtain the latent noise vector with the image manifold’s information and let the generative adversarial network sample the input from this informative noise vector in order to generate the skin lesion images. The use of informative noise allows the GAN to avoid mode collapse and creates faster convergence. To improve the diversity in the generated images, we use another GAN with an auxiliary classifier, which samples the noise vector from a heavy-tailed student t-distribution instead of a random noise Gaussian distribution. The proposed framework was named TED-GAN, with T from the t-distribution and ED from the encoder-decoder network which is part of the solution. The proposed framework could be used in a broad range of areas in medical imaging. We used it here to generate skin lesion images and have obtained an improved classification performance on the skin lesion classification task, rising from 66% average accuracy to 92.5%. The results show that TED-GAN has a better impact on the classification task because of its diverse range of generated images due to the use of a heavy-tailed t-distribution.Bilal AhmadSun JunVasile PaladeQi YouLi MaoMao ZhongjieMDPI AGarticlevariational autoencodergenerative adversarial networksmelanoma detectionskin cancer classificationstudent t-distributionheavy-tailed distributionMedicine (General)R5-920ENDiagnostics, Vol 11, Iss 2147, p 2147 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	variational autoencoder generative adversarial networks melanoma detection skin cancer classification student t-distribution heavy-tailed distribution Medicine (General) R5-920
spellingShingle	variational autoencoder generative adversarial networks melanoma detection skin cancer classification student t-distribution heavy-tailed distribution Medicine (General) R5-920 Bilal Ahmad Sun Jun Vasile Palade Qi You Li Mao Mao Zhongjie Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)
description	Deep learning has gained immense attention from researchers in medicine, especially in medical imaging. The main bottleneck is the unavailability of sufficiently large medical datasets required for the good performance of deep learning models. This paper proposes a new framework consisting of one variational autoencoder (VAE), two generative adversarial networks, and one auxiliary classifier to artificially generate realistic-looking skin lesion images and improve classification performance. We first train the encoder-decoder network to obtain the latent noise vector with the image manifold’s information and let the generative adversarial network sample the input from this informative noise vector in order to generate the skin lesion images. The use of informative noise allows the GAN to avoid mode collapse and creates faster convergence. To improve the diversity in the generated images, we use another GAN with an auxiliary classifier, which samples the noise vector from a heavy-tailed student t-distribution instead of a random noise Gaussian distribution. The proposed framework was named TED-GAN, with T from the t-distribution and ED from the encoder-decoder network which is part of the solution. The proposed framework could be used in a broad range of areas in medical imaging. We used it here to generate skin lesion images and have obtained an improved classification performance on the skin lesion classification task, rising from 66% average accuracy to 92.5%. The results show that TED-GAN has a better impact on the classification task because of its diverse range of generated images due to the use of a heavy-tailed t-distribution.
format	article
author	Bilal Ahmad Sun Jun Vasile Palade Qi You Li Mao Mao Zhongjie
author_facet	Bilal Ahmad Sun Jun Vasile Palade Qi You Li Mao Mao Zhongjie
author_sort	Bilal Ahmad
title	Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)
title_short	Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)
title_full	Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)
title_fullStr	Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)
title_full_unstemmed	Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)
title_sort	improving skin cancer classification using heavy-tailed student t-distribution in generative adversarial networks (ted-gan)
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/3b9db49dbab84d11853d3f1754a2ff87
work_keys_str_mv	AT bilalahmad improvingskincancerclassificationusingheavytailedstudenttdistributioningenerativeadversarialnetworkstedgan AT sunjun improvingskincancerclassificationusingheavytailedstudenttdistributioningenerativeadversarialnetworkstedgan AT vasilepalade improvingskincancerclassificationusingheavytailedstudenttdistributioningenerativeadversarialnetworkstedgan AT qiyou improvingskincancerclassificationusingheavytailedstudenttdistributioningenerativeadversarialnetworkstedgan AT limao improvingskincancerclassificationusingheavytailedstudenttdistributioningenerativeadversarialnetworkstedgan AT maozhongjie improvingskincancerclassificationusingheavytailedstudenttdistributioningenerativeadversarialnetworkstedgan
_version_	1718412464484253696

Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)

Ejemplares similares