Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising

Cross-modal medical imaging techniques are predominantly being used in the clinical suite. The ensemble learning methods using cross-modal medical imaging adds reliability to several medical image analysis tasks. Motivated by the performance of deep learning in several medical imaging tasks, a deep...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Rabia Naseem, Faouzi Alaya Cheikh, Azeddine Beghdadi, Khan Muhammad, Muhammad Sajjad
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	cross-modal guided denoising MRI machine learning siamese network Electronics TK7800-8360
Acceso en línea:	https://doaj.org/article/343a6a49ef9a4ec79202aec6b19fdfb4
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:343a6a49ef9a4ec79202aec6b19fdfb4
record_format	dspace
spelling	oai:doaj.org-article:343a6a49ef9a4ec79202aec6b19fdfb42021-11-25T17:25:21ZCross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising10.3390/electronics102228552079-9292https://doaj.org/article/343a6a49ef9a4ec79202aec6b19fdfb42021-11-01T00:00:00Zhttps://www.mdpi.com/2079-9292/10/22/2855https://doaj.org/toc/2079-9292Cross-modal medical imaging techniques are predominantly being used in the clinical suite. The ensemble learning methods using cross-modal medical imaging adds reliability to several medical image analysis tasks. Motivated by the performance of deep learning in several medical imaging tasks, a deep learning-based denoising method Cross-Modality Guided Denoising Network <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>C</mi><mi>M</mi><mi>G</mi><mi>D</mi><mi>N</mi><mi>e</mi><mi>t</mi></mrow></semantics></math></inline-formula> for removing Rician noise in T1-weighted (T1-w) Magnetic Resonance Images (MRI) is proposed in this paper. <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>C</mi><mi>M</mi><mi>G</mi><mi>D</mi><mi>N</mi><mi>e</mi><mi>t</mi></mrow></semantics></math></inline-formula> uses a guidance image, which is a cross-modal (T2-w) image of better perceptual quality to guide the model in denoising its noisy T1-w counterpart. This cross-modal combination allows the network to exploit complementary information existing in both images and therefore improve the learning capability of the model. The proposed framework consists of two components: Paired Hierarchical Learning (PHL) module and Cross-Modal Assisted Reconstruction (CMAR) module. PHL module uses Siamese network to extract hierarchical features from dual images, which are then combined in a densely connected manner in the CMAR module to finally reconstruct the image. The impact of using registered guidance data is investigated in removing noise as well as retaining structural similarity with the original image. Several experiments were conducted on two publicly available brain imaging datasets available on the IXI database. The quantitative assessment using Peak Signal to noise ratio (PSNR), Structural Similarity Index (SSIM), and Feature Similarity Index (FSIM) demonstrates that the proposed method exhibits 4.7% and 2.3% gain (average), respectively, in SSIM and FSIM values compared to other state-of-the-art denoising methods that do not integrate cross-modal image information in removing various levels of noise.Rabia NaseemFaouzi Alaya CheikhAzeddine BeghdadiKhan MuhammadMuhammad SajjadMDPI AGarticlecross-modalguideddenoisingMRImachine learningsiamese networkElectronicsTK7800-8360ENElectronics, Vol 10, Iss 2855, p 2855 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	cross-modal guided denoising MRI machine learning siamese network Electronics TK7800-8360
spellingShingle	cross-modal guided denoising MRI machine learning siamese network Electronics TK7800-8360 Rabia Naseem Faouzi Alaya Cheikh Azeddine Beghdadi Khan Muhammad Muhammad Sajjad Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising
description	Cross-modal medical imaging techniques are predominantly being used in the clinical suite. The ensemble learning methods using cross-modal medical imaging adds reliability to several medical image analysis tasks. Motivated by the performance of deep learning in several medical imaging tasks, a deep learning-based denoising method Cross-Modality Guided Denoising Network <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>C</mi><mi>M</mi><mi>G</mi><mi>D</mi><mi>N</mi><mi>e</mi><mi>t</mi></mrow></semantics></math></inline-formula> for removing Rician noise in T1-weighted (T1-w) Magnetic Resonance Images (MRI) is proposed in this paper. <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>C</mi><mi>M</mi><mi>G</mi><mi>D</mi><mi>N</mi><mi>e</mi><mi>t</mi></mrow></semantics></math></inline-formula> uses a guidance image, which is a cross-modal (T2-w) image of better perceptual quality to guide the model in denoising its noisy T1-w counterpart. This cross-modal combination allows the network to exploit complementary information existing in both images and therefore improve the learning capability of the model. The proposed framework consists of two components: Paired Hierarchical Learning (PHL) module and Cross-Modal Assisted Reconstruction (CMAR) module. PHL module uses Siamese network to extract hierarchical features from dual images, which are then combined in a densely connected manner in the CMAR module to finally reconstruct the image. The impact of using registered guidance data is investigated in removing noise as well as retaining structural similarity with the original image. Several experiments were conducted on two publicly available brain imaging datasets available on the IXI database. The quantitative assessment using Peak Signal to noise ratio (PSNR), Structural Similarity Index (SSIM), and Feature Similarity Index (FSIM) demonstrates that the proposed method exhibits 4.7% and 2.3% gain (average), respectively, in SSIM and FSIM values compared to other state-of-the-art denoising methods that do not integrate cross-modal image information in removing various levels of noise.
format	article
author	Rabia Naseem Faouzi Alaya Cheikh Azeddine Beghdadi Khan Muhammad Muhammad Sajjad
author_facet	Rabia Naseem Faouzi Alaya Cheikh Azeddine Beghdadi Khan Muhammad Muhammad Sajjad
author_sort	Rabia Naseem
title	Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising
title_short	Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising
title_full	Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising
title_fullStr	Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising
title_full_unstemmed	Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising
title_sort	cross-modal guidance assisted hierarchical learning based siamese network for mr image denoising
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/343a6a49ef9a4ec79202aec6b19fdfb4
work_keys_str_mv	AT rabianaseem crossmodalguidanceassistedhierarchicallearningbasedsiamesenetworkformrimagedenoising AT faouzialayacheikh crossmodalguidanceassistedhierarchicallearningbasedsiamesenetworkformrimagedenoising AT azeddinebeghdadi crossmodalguidanceassistedhierarchicallearningbasedsiamesenetworkformrimagedenoising AT khanmuhammad crossmodalguidanceassistedhierarchicallearningbasedsiamesenetworkformrimagedenoising AT muhammadsajjad crossmodalguidanceassistedhierarchicallearningbasedsiamesenetworkformrimagedenoising
_version_	1718412343010918400

Cross-Modal Guidance Assisted Hierarchical Learning Based Siamese Network for MR Image Denoising

Ejemplares similares