Loading…

A deep learning-based reconstruction approach for accelerated magnetic resonance image of the knee with compressed sense: evaluation in healthy volunteers

To evaluate the feasibility of combining compressed sense (CS) with a newly developed deep learning-based algorithm (CS-AI) using convolutional neural networks to accelerate 2D MRI of the knee. In this prospective study, 20 healthy volunteers were scanned with a 3T MRI scanner. All subjects received...

Full description

Saved in:
Bibliographic Details
Published in:British journal of radiology 2023-06, Vol.96 (1146), p.20220074-20220074
Main Authors: Iuga, Andra-Iza, Rauen, Philip Santiago, Siedek, Florian, Große-Hokamp, Nils, Sonnabend, Kristina, Maintz, David, Lennartz, Simon, Bratke, Grischa
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513
cites cdi_FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513
container_end_page 20220074
container_issue 1146
container_start_page 20220074
container_title British journal of radiology
container_volume 96
creator Iuga, Andra-Iza
Rauen, Philip Santiago
Siedek, Florian
Große-Hokamp, Nils
Sonnabend, Kristina
Maintz, David
Lennartz, Simon
Bratke, Grischa
description To evaluate the feasibility of combining compressed sense (CS) with a newly developed deep learning-based algorithm (CS-AI) using convolutional neural networks to accelerate 2D MRI of the knee. In this prospective study, 20 healthy volunteers were scanned with a 3T MRI scanner. All subjects received a fat-saturated sagittal 2D proton density reference sequence without acceleration and four additional acquisitions with different acceleration levels: 2, 3, 4 and 6. All sequences were reconstructed with the conventional CS and a new CS-AI algorithm. Two independent, blinded readers rated all images by seven criteria (overall image impression, visible artifacts, delineation of anterior ligament, posterior ligament, menisci, cartilage, and bone) using a 5-point Likert scale. Signal- and contrast-to-noise ratios were calculated. Subjective ratings and quantitative metrics were compared between CS and CS-AI with similar acceleration levels and between all CS/CS-AI images and the non-accelerated reference sequence. Friedman and Dunn´s multiple comparison tests were used for subjective, ANOVA and the Tukey Kramer test for quantitative metrics. Conventional CS images at the lowest acceleration level (CS2) were already rated significantly lower than reference for 6/7 criteria. CS-AI images maintained similar image quality to the reference up to CS-AI three for all criteria, which would allow for a reduction in scan time of 64% with unchanged image quality compared to the unaccelerated sequence. SNR and CNR were significantly higher for all CS-AI reconstructions compared to CS (all < 0.05). AI-based image reconstruction showed higher image quality than CS for 2D knee imaging. Its implementation in the clinical routine yields the potential for faster MRI acquisition but needs further validation in non-healthy study subjects. Combining compressed SENSE with a newly developed deep learning-based algorithm using convolutional neural networks allows a 64% reduction in scan time for 2D imaging of the knee. Implementation of the new deep learning-based algorithm in clinical routine in near future should enable better image quality/resolution with constant scan time, or reduced acquisition times while maintaining diagnostic quality.
doi_str_mv 10.1259/bjr.20220074
format article
fullrecord <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_10230385</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2805030061</sourcerecordid><originalsourceid>FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513</originalsourceid><addsrcrecordid>eNpVkUtv1TAQRi0EopfCjjXykgUpYztPNqiqeEmV2LQSO2viTG5ccu1gO7fqX-mvxZc-VFaWx8dnxv4YeyvgRMiq-9hfhRMJUgI05TO2EU3ZFm0Lv56zDeRaIWRbHbFXMV4dtlUHL9mRaqCtoWk27PaUD0QLnwmDs25b9Bhp4IGMdzGF1STrHcdlCR7NxEcfOBpDMwVMmdvh1lGyJl-I3qEzxG2uEfcjTxPx346IX9s0ceN3S4YO8kgu0idOe5xX_Oe3jk-Ec5pu-N7Pq0tEIb5mL0acI725X4_Z5dcvF2ffi_Of336cnZ4XRrVVKkxveolGDS1W_dDUEhRKUZUtlF0N46jKWqieumYwVIlukDh0oh8bQ7USQyXUMft8513WfkeZcingrJeQXxJutEer_z9xdtJbv9cCpII8Qza8vzcE_2elmPTOxvxJMzrya9SyhQoUQH1o9uEONcHHGGh87CNAH_LUOU_9kGfG3z2d7RF-CFD9BbQ8oPo</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2805030061</pqid></control><display><type>article</type><title>A deep learning-based reconstruction approach for accelerated magnetic resonance image of the knee with compressed sense: evaluation in healthy volunteers</title><source>Oxford Journals Online</source><source>Alma/SFX Local Collection</source><creator>Iuga, Andra-Iza ; Rauen, Philip Santiago ; Siedek, Florian ; Große-Hokamp, Nils ; Sonnabend, Kristina ; Maintz, David ; Lennartz, Simon ; Bratke, Grischa</creator><creatorcontrib>Iuga, Andra-Iza ; Rauen, Philip Santiago ; Siedek, Florian ; Große-Hokamp, Nils ; Sonnabend, Kristina ; Maintz, David ; Lennartz, Simon ; Bratke, Grischa</creatorcontrib><description>To evaluate the feasibility of combining compressed sense (CS) with a newly developed deep learning-based algorithm (CS-AI) using convolutional neural networks to accelerate 2D MRI of the knee. In this prospective study, 20 healthy volunteers were scanned with a 3T MRI scanner. All subjects received a fat-saturated sagittal 2D proton density reference sequence without acceleration and four additional acquisitions with different acceleration levels: 2, 3, 4 and 6. All sequences were reconstructed with the conventional CS and a new CS-AI algorithm. Two independent, blinded readers rated all images by seven criteria (overall image impression, visible artifacts, delineation of anterior ligament, posterior ligament, menisci, cartilage, and bone) using a 5-point Likert scale. Signal- and contrast-to-noise ratios were calculated. Subjective ratings and quantitative metrics were compared between CS and CS-AI with similar acceleration levels and between all CS/CS-AI images and the non-accelerated reference sequence. Friedman and Dunn´s multiple comparison tests were used for subjective, ANOVA and the Tukey Kramer test for quantitative metrics. Conventional CS images at the lowest acceleration level (CS2) were already rated significantly lower than reference for 6/7 criteria. CS-AI images maintained similar image quality to the reference up to CS-AI three for all criteria, which would allow for a reduction in scan time of 64% with unchanged image quality compared to the unaccelerated sequence. SNR and CNR were significantly higher for all CS-AI reconstructions compared to CS (all &lt; 0.05). AI-based image reconstruction showed higher image quality than CS for 2D knee imaging. Its implementation in the clinical routine yields the potential for faster MRI acquisition but needs further validation in non-healthy study subjects. Combining compressed SENSE with a newly developed deep learning-based algorithm using convolutional neural networks allows a 64% reduction in scan time for 2D imaging of the knee. Implementation of the new deep learning-based algorithm in clinical routine in near future should enable better image quality/resolution with constant scan time, or reduced acquisition times while maintaining diagnostic quality.</description><identifier>ISSN: 0007-1285</identifier><identifier>EISSN: 1748-880X</identifier><identifier>DOI: 10.1259/bjr.20220074</identifier><identifier>PMID: 37086077</identifier><language>eng</language><publisher>England: The British Institute of Radiology</publisher><subject>Deep Learning ; Healthy Volunteers ; Humans ; Image Processing, Computer-Assisted - methods ; Imaging, Three-Dimensional - methods ; Knee Joint - diagnostic imaging ; Magnetic Resonance Imaging - methods ; Prospective Studies</subject><ispartof>British journal of radiology, 2023-06, Vol.96 (1146), p.20220074-20220074</ispartof><rights>2023 The Authors. Published by the British Institute of Radiology 2023 The Authors</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513</citedby><cites>FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513</cites><orcidid>0000-0002-3694-0235</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,780,784,885,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/37086077$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Iuga, Andra-Iza</creatorcontrib><creatorcontrib>Rauen, Philip Santiago</creatorcontrib><creatorcontrib>Siedek, Florian</creatorcontrib><creatorcontrib>Große-Hokamp, Nils</creatorcontrib><creatorcontrib>Sonnabend, Kristina</creatorcontrib><creatorcontrib>Maintz, David</creatorcontrib><creatorcontrib>Lennartz, Simon</creatorcontrib><creatorcontrib>Bratke, Grischa</creatorcontrib><title>A deep learning-based reconstruction approach for accelerated magnetic resonance image of the knee with compressed sense: evaluation in healthy volunteers</title><title>British journal of radiology</title><addtitle>Br J Radiol</addtitle><description>To evaluate the feasibility of combining compressed sense (CS) with a newly developed deep learning-based algorithm (CS-AI) using convolutional neural networks to accelerate 2D MRI of the knee. In this prospective study, 20 healthy volunteers were scanned with a 3T MRI scanner. All subjects received a fat-saturated sagittal 2D proton density reference sequence without acceleration and four additional acquisitions with different acceleration levels: 2, 3, 4 and 6. All sequences were reconstructed with the conventional CS and a new CS-AI algorithm. Two independent, blinded readers rated all images by seven criteria (overall image impression, visible artifacts, delineation of anterior ligament, posterior ligament, menisci, cartilage, and bone) using a 5-point Likert scale. Signal- and contrast-to-noise ratios were calculated. Subjective ratings and quantitative metrics were compared between CS and CS-AI with similar acceleration levels and between all CS/CS-AI images and the non-accelerated reference sequence. Friedman and Dunn´s multiple comparison tests were used for subjective, ANOVA and the Tukey Kramer test for quantitative metrics. Conventional CS images at the lowest acceleration level (CS2) were already rated significantly lower than reference for 6/7 criteria. CS-AI images maintained similar image quality to the reference up to CS-AI three for all criteria, which would allow for a reduction in scan time of 64% with unchanged image quality compared to the unaccelerated sequence. SNR and CNR were significantly higher for all CS-AI reconstructions compared to CS (all &lt; 0.05). AI-based image reconstruction showed higher image quality than CS for 2D knee imaging. Its implementation in the clinical routine yields the potential for faster MRI acquisition but needs further validation in non-healthy study subjects. Combining compressed SENSE with a newly developed deep learning-based algorithm using convolutional neural networks allows a 64% reduction in scan time for 2D imaging of the knee. Implementation of the new deep learning-based algorithm in clinical routine in near future should enable better image quality/resolution with constant scan time, or reduced acquisition times while maintaining diagnostic quality.</description><subject>Deep Learning</subject><subject>Healthy Volunteers</subject><subject>Humans</subject><subject>Image Processing, Computer-Assisted - methods</subject><subject>Imaging, Three-Dimensional - methods</subject><subject>Knee Joint - diagnostic imaging</subject><subject>Magnetic Resonance Imaging - methods</subject><subject>Prospective Studies</subject><issn>0007-1285</issn><issn>1748-880X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNpVkUtv1TAQRi0EopfCjjXykgUpYztPNqiqeEmV2LQSO2viTG5ccu1gO7fqX-mvxZc-VFaWx8dnxv4YeyvgRMiq-9hfhRMJUgI05TO2EU3ZFm0Lv56zDeRaIWRbHbFXMV4dtlUHL9mRaqCtoWk27PaUD0QLnwmDs25b9Bhp4IGMdzGF1STrHcdlCR7NxEcfOBpDMwVMmdvh1lGyJl-I3qEzxG2uEfcjTxPx346IX9s0ceN3S4YO8kgu0idOe5xX_Oe3jk-Ec5pu-N7Pq0tEIb5mL0acI725X4_Z5dcvF2ffi_Of336cnZ4XRrVVKkxveolGDS1W_dDUEhRKUZUtlF0N46jKWqieumYwVIlukDh0oh8bQ7USQyXUMft8513WfkeZcingrJeQXxJutEer_z9xdtJbv9cCpII8Qza8vzcE_2elmPTOxvxJMzrya9SyhQoUQH1o9uEONcHHGGh87CNAH_LUOU_9kGfG3z2d7RF-CFD9BbQ8oPo</recordid><startdate>20230601</startdate><enddate>20230601</enddate><creator>Iuga, Andra-Iza</creator><creator>Rauen, Philip Santiago</creator><creator>Siedek, Florian</creator><creator>Große-Hokamp, Nils</creator><creator>Sonnabend, Kristina</creator><creator>Maintz, David</creator><creator>Lennartz, Simon</creator><creator>Bratke, Grischa</creator><general>The British Institute of Radiology</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-3694-0235</orcidid></search><sort><creationdate>20230601</creationdate><title>A deep learning-based reconstruction approach for accelerated magnetic resonance image of the knee with compressed sense: evaluation in healthy volunteers</title><author>Iuga, Andra-Iza ; Rauen, Philip Santiago ; Siedek, Florian ; Große-Hokamp, Nils ; Sonnabend, Kristina ; Maintz, David ; Lennartz, Simon ; Bratke, Grischa</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Deep Learning</topic><topic>Healthy Volunteers</topic><topic>Humans</topic><topic>Image Processing, Computer-Assisted - methods</topic><topic>Imaging, Three-Dimensional - methods</topic><topic>Knee Joint - diagnostic imaging</topic><topic>Magnetic Resonance Imaging - methods</topic><topic>Prospective Studies</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Iuga, Andra-Iza</creatorcontrib><creatorcontrib>Rauen, Philip Santiago</creatorcontrib><creatorcontrib>Siedek, Florian</creatorcontrib><creatorcontrib>Große-Hokamp, Nils</creatorcontrib><creatorcontrib>Sonnabend, Kristina</creatorcontrib><creatorcontrib>Maintz, David</creatorcontrib><creatorcontrib>Lennartz, Simon</creatorcontrib><creatorcontrib>Bratke, Grischa</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>British journal of radiology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Iuga, Andra-Iza</au><au>Rauen, Philip Santiago</au><au>Siedek, Florian</au><au>Große-Hokamp, Nils</au><au>Sonnabend, Kristina</au><au>Maintz, David</au><au>Lennartz, Simon</au><au>Bratke, Grischa</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A deep learning-based reconstruction approach for accelerated magnetic resonance image of the knee with compressed sense: evaluation in healthy volunteers</atitle><jtitle>British journal of radiology</jtitle><addtitle>Br J Radiol</addtitle><date>2023-06-01</date><risdate>2023</risdate><volume>96</volume><issue>1146</issue><spage>20220074</spage><epage>20220074</epage><pages>20220074-20220074</pages><issn>0007-1285</issn><eissn>1748-880X</eissn><abstract>To evaluate the feasibility of combining compressed sense (CS) with a newly developed deep learning-based algorithm (CS-AI) using convolutional neural networks to accelerate 2D MRI of the knee. In this prospective study, 20 healthy volunteers were scanned with a 3T MRI scanner. All subjects received a fat-saturated sagittal 2D proton density reference sequence without acceleration and four additional acquisitions with different acceleration levels: 2, 3, 4 and 6. All sequences were reconstructed with the conventional CS and a new CS-AI algorithm. Two independent, blinded readers rated all images by seven criteria (overall image impression, visible artifacts, delineation of anterior ligament, posterior ligament, menisci, cartilage, and bone) using a 5-point Likert scale. Signal- and contrast-to-noise ratios were calculated. Subjective ratings and quantitative metrics were compared between CS and CS-AI with similar acceleration levels and between all CS/CS-AI images and the non-accelerated reference sequence. Friedman and Dunn´s multiple comparison tests were used for subjective, ANOVA and the Tukey Kramer test for quantitative metrics. Conventional CS images at the lowest acceleration level (CS2) were already rated significantly lower than reference for 6/7 criteria. CS-AI images maintained similar image quality to the reference up to CS-AI three for all criteria, which would allow for a reduction in scan time of 64% with unchanged image quality compared to the unaccelerated sequence. SNR and CNR were significantly higher for all CS-AI reconstructions compared to CS (all &lt; 0.05). AI-based image reconstruction showed higher image quality than CS for 2D knee imaging. Its implementation in the clinical routine yields the potential for faster MRI acquisition but needs further validation in non-healthy study subjects. Combining compressed SENSE with a newly developed deep learning-based algorithm using convolutional neural networks allows a 64% reduction in scan time for 2D imaging of the knee. Implementation of the new deep learning-based algorithm in clinical routine in near future should enable better image quality/resolution with constant scan time, or reduced acquisition times while maintaining diagnostic quality.</abstract><cop>England</cop><pub>The British Institute of Radiology</pub><pmid>37086077</pmid><doi>10.1259/bjr.20220074</doi><tpages>1</tpages><orcidid>https://orcid.org/0000-0002-3694-0235</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0007-1285
ispartof British journal of radiology, 2023-06, Vol.96 (1146), p.20220074-20220074
issn 0007-1285
1748-880X
language eng
recordid cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_10230385
source Oxford Journals Online; Alma/SFX Local Collection
subjects Deep Learning
Healthy Volunteers
Humans
Image Processing, Computer-Assisted - methods
Imaging, Three-Dimensional - methods
Knee Joint - diagnostic imaging
Magnetic Resonance Imaging - methods
Prospective Studies
title A deep learning-based reconstruction approach for accelerated magnetic resonance image of the knee with compressed sense: evaluation in healthy volunteers
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T19%3A56%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20deep%20learning-based%20reconstruction%20approach%20for%20accelerated%20magnetic%20resonance%20image%20of%20the%20knee%20with%20compressed%20sense:%20evaluation%20in%20healthy%20volunteers&rft.jtitle=British%20journal%20of%20radiology&rft.au=Iuga,%20Andra-Iza&rft.date=2023-06-01&rft.volume=96&rft.issue=1146&rft.spage=20220074&rft.epage=20220074&rft.pages=20220074-20220074&rft.issn=0007-1285&rft.eissn=1748-880X&rft_id=info:doi/10.1259/bjr.20220074&rft_dat=%3Cproquest_pubme%3E2805030061%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c385t-cbcb2ac3d8a5bd76203a2154804960ff34613be97dce519d2ad91bf7ce631d513%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2805030061&rft_id=info:pmid/37086077&rfr_iscdi=true