A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction

Abstract Grading individual knee osteoarthritis (OA) features is a fine-grained knee OA severity assessment. Existing methods ignore following problems: (1) more accurately located knee joints benefit subsequent grades prediction; (2) they do not consider knee joints’ symmetry and semantic informati...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Kang Wang, Xin Niu, Yong Dou, Dongxing Xie, Tuo Yang
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/1af309dead77435b9a09277994c55141
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1af309dead77435b9a09277994c55141
record_format	dspace
spelling	oai:doaj.org-article:1af309dead77435b9a09277994c551412021-12-02T18:51:52ZA siamese network with adaptive gated feature fusion for individual knee OA features grades prediction10.1038/s41598-021-96240-82045-2322https://doaj.org/article/1af309dead77435b9a09277994c551412021-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-96240-8https://doaj.org/toc/2045-2322Abstract Grading individual knee osteoarthritis (OA) features is a fine-grained knee OA severity assessment. Existing methods ignore following problems: (1) more accurately located knee joints benefit subsequent grades prediction; (2) they do not consider knee joints’ symmetry and semantic information, which help to improve grades prediction performance. To this end, we propose a SE-ResNext50-32x4d-based Siamese network with adaptive gated feature fusion method to simultaneously assess eight tasks. In our method, two cascaded small convolution neural networks are designed to locate more accurate knee joints. Detected knee joints are further cropped and split into left and right patches via their symmetry, which are fed into SE-ResNext50-32x4d-based Siamese network with shared weights, extracting more detailed knee features. The adaptive gated feature fusion method is used to capture richer semantic information for better feature representation here. Meanwhile, knee OA/non-knee OA classification task is added, helping extract richer features. We specially introduce a new evaluation metric (top±1 accuracy) aiming to measure model performance with ambiguous data labels. Our model is evaluated on two public datasets: OAI and MOST datasets, achieving the state-of-the-art results comparing to competing approaches. It has the potential to be a tool to assist clinical decision making.Kang WangXin NiuYong DouDongxing XieTuo YangNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Kang Wang Xin Niu Yong Dou Dongxing Xie Tuo Yang A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction
description	Abstract Grading individual knee osteoarthritis (OA) features is a fine-grained knee OA severity assessment. Existing methods ignore following problems: (1) more accurately located knee joints benefit subsequent grades prediction; (2) they do not consider knee joints’ symmetry and semantic information, which help to improve grades prediction performance. To this end, we propose a SE-ResNext50-32x4d-based Siamese network with adaptive gated feature fusion method to simultaneously assess eight tasks. In our method, two cascaded small convolution neural networks are designed to locate more accurate knee joints. Detected knee joints are further cropped and split into left and right patches via their symmetry, which are fed into SE-ResNext50-32x4d-based Siamese network with shared weights, extracting more detailed knee features. The adaptive gated feature fusion method is used to capture richer semantic information for better feature representation here. Meanwhile, knee OA/non-knee OA classification task is added, helping extract richer features. We specially introduce a new evaluation metric (top±1 accuracy) aiming to measure model performance with ambiguous data labels. Our model is evaluated on two public datasets: OAI and MOST datasets, achieving the state-of-the-art results comparing to competing approaches. It has the potential to be a tool to assist clinical decision making.
format	article
author	Kang Wang Xin Niu Yong Dou Dongxing Xie Tuo Yang
author_facet	Kang Wang Xin Niu Yong Dou Dongxing Xie Tuo Yang
author_sort	Kang Wang
title	A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction
title_short	A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction
title_full	A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction
title_fullStr	A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction
title_full_unstemmed	A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction
title_sort	siamese network with adaptive gated feature fusion for individual knee oa features grades prediction
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/1af309dead77435b9a09277994c55141
work_keys_str_mv	AT kangwang asiamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT xinniu asiamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT yongdou asiamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT dongxingxie asiamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT tuoyang asiamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT kangwang siamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT xinniu siamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT yongdou siamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT dongxingxie siamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction AT tuoyang siamesenetworkwithadaptivegatedfeaturefusionforindividualkneeoafeaturesgradesprediction
_version_	1718377373012852736

A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction

Ejemplares similares