AI-based localization and classification of skin disease with erythema

Abstract Although computer-aided diagnosis (CAD) is used to improve the quality of diagnosis in various medical fields such as mammography and colonography, it is not used in dermatology, where noninvasive screening tests are performed only with the naked eye, and avoidable inaccuracies may exist. T...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Ha Min Son, Wooho Jeon, Jinhyun Kim, Chan Yeong Heo, Hye Jin Yoon, Ji-Ung Park, Tai-Myoung Chung
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/2ce2d815162a4b5bb6ff7d8de7f7f5f7
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:2ce2d815162a4b5bb6ff7d8de7f7f5f7
record_format	dspace
spelling	oai:doaj.org-article:2ce2d815162a4b5bb6ff7d8de7f7f5f72021-12-02T13:20:22ZAI-based localization and classification of skin disease with erythema10.1038/s41598-021-84593-z2045-2322https://doaj.org/article/2ce2d815162a4b5bb6ff7d8de7f7f5f72021-03-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-84593-zhttps://doaj.org/toc/2045-2322Abstract Although computer-aided diagnosis (CAD) is used to improve the quality of diagnosis in various medical fields such as mammography and colonography, it is not used in dermatology, where noninvasive screening tests are performed only with the naked eye, and avoidable inaccuracies may exist. This study shows that CAD may also be a viable option in dermatology by presenting a novel method to sequentially combine accurate segmentation and classification models. Given an image of the skin, we decompose the image to normalize and extract high-level features. Using a neural network-based segmentation model to create a segmented map of the image, we then cluster sections of abnormal skin and pass this information to a classification model. We classify each cluster into different common skin diseases using another neural network model. Our segmentation model achieves better performance compared to previous studies, and also achieves a near-perfect sensitivity score in unfavorable conditions. Our classification model is more accurate than a baseline model trained without segmentation, while also being able to classify multiple diseases within a single image. This improved performance may be sufficient to use CAD in the field of dermatology.Ha Min SonWooho JeonJinhyun KimChan Yeong HeoHye Jin YoonJi-Ung ParkTai-Myoung ChungNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-14 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Ha Min Son Wooho Jeon Jinhyun Kim Chan Yeong Heo Hye Jin Yoon Ji-Ung Park Tai-Myoung Chung AI-based localization and classification of skin disease with erythema
description	Abstract Although computer-aided diagnosis (CAD) is used to improve the quality of diagnosis in various medical fields such as mammography and colonography, it is not used in dermatology, where noninvasive screening tests are performed only with the naked eye, and avoidable inaccuracies may exist. This study shows that CAD may also be a viable option in dermatology by presenting a novel method to sequentially combine accurate segmentation and classification models. Given an image of the skin, we decompose the image to normalize and extract high-level features. Using a neural network-based segmentation model to create a segmented map of the image, we then cluster sections of abnormal skin and pass this information to a classification model. We classify each cluster into different common skin diseases using another neural network model. Our segmentation model achieves better performance compared to previous studies, and also achieves a near-perfect sensitivity score in unfavorable conditions. Our classification model is more accurate than a baseline model trained without segmentation, while also being able to classify multiple diseases within a single image. This improved performance may be sufficient to use CAD in the field of dermatology.
format	article
author	Ha Min Son Wooho Jeon Jinhyun Kim Chan Yeong Heo Hye Jin Yoon Ji-Ung Park Tai-Myoung Chung
author_facet	Ha Min Son Wooho Jeon Jinhyun Kim Chan Yeong Heo Hye Jin Yoon Ji-Ung Park Tai-Myoung Chung
author_sort	Ha Min Son
title	AI-based localization and classification of skin disease with erythema
title_short	AI-based localization and classification of skin disease with erythema
title_full	AI-based localization and classification of skin disease with erythema
title_fullStr	AI-based localization and classification of skin disease with erythema
title_full_unstemmed	AI-based localization and classification of skin disease with erythema
title_sort	ai-based localization and classification of skin disease with erythema
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/2ce2d815162a4b5bb6ff7d8de7f7f5f7
work_keys_str_mv	AT haminson aibasedlocalizationandclassificationofskindiseasewitherythema AT woohojeon aibasedlocalizationandclassificationofskindiseasewitherythema AT jinhyunkim aibasedlocalizationandclassificationofskindiseasewitherythema AT chanyeongheo aibasedlocalizationandclassificationofskindiseasewitherythema AT hyejinyoon aibasedlocalizationandclassificationofskindiseasewitherythema AT jiungpark aibasedlocalizationandclassificationofskindiseasewitherythema AT taimyoungchung aibasedlocalizationandclassificationofskindiseasewitherythema
_version_	1718393199833120768

AI-based localization and classification of skin disease with erythema

Ejemplares similares