Trabelsi Amira
Keywords
- Antenna RF
- MR Physics / MR Method developments
- Radio Frequency
- Ultra-high field MRI
Publications :
180164
Trabelsi
1
harvard-cite-them-right-no-et-al
50
date
desc
year
5035
https://crmbm.univ-amu.fr/wp-content/plugins/zotpress/
%7B%22status%22%3A%22success%22%2C%22updateneeded%22%3Afalse%2C%22instance%22%3Afalse%2C%22meta%22%3A%7B%22request_last%22%3A0%2C%22request_next%22%3A0%2C%22used_cache%22%3Atrue%7D%2C%22data%22%3A%5B%7B%22key%22%3A%22T9ZKI5F9%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Martin%20et%20al.%22%2C%22parsedDate%22%3A%222025-09-08%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BMartin%2C%20S.%2C%20Trabelsi%2C%20A.%2C%20Guye%2C%20M.%2C%20Dubois%2C%20M.%2C%20Abdeddaim%2C%20R.%2C%20Bendahan%2C%20D.%20and%20Andr%26%23xE9%3B%2C%20R.%20%282025%29%20%26%23x201C%3BOptimized%20reconstruction%20of%20undersampled%20Dixon%20sequences%20using%20new%20memory-efficient%20unrolled%20deep%20neural%20networks%3A%20HalfVarNet%20and%20HalfDIRCN%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BMagnetic%20Resonance%20in%20Medicine%26lt%3B%5C%2Fi%26gt%3B%20%5BPreprint%5D.%20Available%20at%3A%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1002%5C%2Fmrm.70070%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1002%5C%2Fmrm.70070%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Optimized%20reconstruction%20of%20undersampled%20Dixon%20sequences%20using%20new%20memory-efficient%20unrolled%20deep%20neural%20networks%3A%20HalfVarNet%20and%20HalfDIRCN%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sandra%22%2C%22lastName%22%3A%22Martin%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amira%22%2C%22lastName%22%3A%22Trabelsi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Maxime%22%2C%22lastName%22%3A%22Guye%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marc%22%2C%22lastName%22%3A%22Dubois%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Redha%22%2C%22lastName%22%3A%22Abdeddaim%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22David%22%2C%22lastName%22%3A%22Bendahan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22R%5Cu00e9mi%22%2C%22lastName%22%3A%22Andr%5Cu00e9%22%7D%5D%2C%22abstractNote%22%3A%22PURPOSE%3A%20Fat%20fraction%20%28FF%29%20quantification%20in%20individual%20muscles%20using%20quantitative%20MRI%20is%20of%20major%20importance%20for%20monitoring%20disease%20progression%20and%20assessing%20disease%20severity%20in%20neuromuscular%20diseases.%20Undersampling%20of%20MRI%20acquisitions%20is%20commonly%20used%20to%20reduce%20scanning%20time.%20The%20present%20paper%20introduces%20novel%20unrolled%20neural%20networks%20for%20the%20reconstruction%20of%20undersampled%20MRI%20acquisitions.%20These%20networks%20are%20designed%20with%20the%20aim%20of%20maintaining%20accurate%20FF%20quantification%20while%20reducing%20reconstruction%20time%20and%20memory%20usage.%5CnMETHODS%3A%20The%20proposed%20approach%20relies%20on%20a%20combination%20of%20a%20simplified%20architecture%20%28Half%20U-Net%29%20with%20unrolled%20networks%20that%20achieved%20high%20performance%20in%20the%20well-known%20FastMRI%20challenge%20%28variational%20network%20%5BVarNet%5D%20and%20densely%20interconnected%20residual%20cascading%20network%20%5BDIRCN%5D%29.%20The%20algorithms%20were%20trained%20and%20evaluated%20using%203D%20MRI%20Dixon%20acquisitions%20of%20the%20thigh%20from%20controls%20and%20patients%20with%20neuromuscular%20diseases.%20The%20study%20was%20performed%20by%20applying%20a%20retrospective%20undersampling%20with%20acceleration%20factors%20of%204%20and%208.%20Reconstructed%20images%20were%20used%20to%20computed%20FF%20maps.%5CnRESULTS%3A%20Results%20disclose%20that%20the%20novel%20unrolled%20neural%20networks%20were%20able%20to%20maintain%20reconstruction%2C%20biomarker%20assessment%2C%20and%20segmentation%20quality%20while%20reducing%20memory%20usage%20by%2024%25%20to%2016%25%20and%20reducing%20reconstruction%20time%20from%2021%25%20to%2017%25.%20Using%20an%20acceleration%20factor%20of%208%2C%20the%20proposed%20algorithms%2C%20HalfVarNet%20and%20HalfDIRCN%2C%20achieved%20structural%20similarity%20index%20%28SSIM%29%20scores%20of%2093.76%5Cu2009%5Cu00b1%5Cu20090.38%20and%2094.95%5Cu2009%5Cu00b1%5Cu20090.32%2C%20mean%20squared%20error%20%28MSE%29%20values%20of%2012.76%5Cu2009%5Cu00b1%5Cu20091.08%5Cu2009%5Cu00d7%5Cu200910-2%20and%2010.25%5Cu2009%5Cu00b1%5Cu20090.87%20%5Cu00d7%5Cu200910-2%2C%20and%20a%20relative%20FF%20quadratic%20error%20of%200.23%5Cu2009%5Cu00b1%5Cu20090.02%25%20and%200.17%5Cu2009%5Cu00b1%5Cu20090.02%25%2C%20respectively.%5CnCONCLUSION%3A%20The%20proposed%20method%20enables%20time%20and%20memory-efficient%20reconstruction%20of%20undersampled%203D%20MRI%20data%2C%20supporting%20its%20potential%20for%20clinical%20application.%22%2C%22date%22%3A%222025-09-08%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1002%5C%2Fmrm.70070%22%2C%22ISSN%22%3A%221522-2594%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22NW6SZQBR%22%2C%22XTA6KS7L%22%5D%2C%22dateModified%22%3A%222025-09-12T07%3A58%3A05Z%22%7D%7D%2C%7B%22key%22%3A%22PY4E2DQ4%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Martin%20et%20al.%22%2C%22parsedDate%22%3A%222025-01-11%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BMartin%2C%20S.%2C%20Andr%26%23xE9%3B%2C%20R.%2C%20Trabelsi%2C%20A.%2C%20Michel%2C%20C.P.%2C%20Fortanier%2C%20E.%2C%20Attarian%2C%20S.%2C%20Guye%2C%20M.%2C%20Dubois%2C%20M.%2C%20Abdeddaim%2C%20R.%20and%20Bendahan%2C%20D.%20%282025%29%20%26%23x201C%3BImportance%20of%20neural%20network%20complexity%20for%20the%20automatic%20segmentation%20of%20individual%20thigh%20muscles%20in%20MRI%20images%20from%20patients%20with%20neuromuscular%20diseases%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BMagma%20%28New%20York%2C%20N.Y.%29%26lt%3B%5C%2Fi%26gt%3B%20%5BPreprint%5D.%20Available%20at%3A%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1007%5C%2Fs10334-024-01221-3%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1007%5C%2Fs10334-024-01221-3%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Importance%20of%20neural%20network%20complexity%20for%20the%20automatic%20segmentation%20of%20individual%20thigh%20muscles%20in%20MRI%20images%20from%20patients%20with%20neuromuscular%20diseases%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sandra%22%2C%22lastName%22%3A%22Martin%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22R%5Cu00e9mi%22%2C%22lastName%22%3A%22Andr%5Cu00e9%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amira%22%2C%22lastName%22%3A%22Trabelsi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Constance%20P.%22%2C%22lastName%22%3A%22Michel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Etienne%22%2C%22lastName%22%3A%22Fortanier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shahram%22%2C%22lastName%22%3A%22Attarian%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Maxime%22%2C%22lastName%22%3A%22Guye%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marc%22%2C%22lastName%22%3A%22Dubois%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Redha%22%2C%22lastName%22%3A%22Abdeddaim%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22David%22%2C%22lastName%22%3A%22Bendahan%22%7D%5D%2C%22abstractNote%22%3A%22OBJECTIVE%3A%20Segmentation%20of%20individual%20thigh%20muscles%20in%20MRI%20images%20is%20essential%20for%20monitoring%20neuromuscular%20diseases%20and%20quantifying%20relevant%20biomarkers%20such%20as%20fat%20fraction%20%28FF%29.%20Deep%20learning%20approaches%20such%20as%20U-Net%20have%20demonstrated%20effectiveness%20in%20this%20field.%20However%2C%20the%20impact%20of%20reducing%20neural%20network%20complexity%20remains%20unexplored%20in%20the%20FF%20quantification%20in%20individual%20muscles.%5CnMATERIAL%20AND%20METHODS%3A%20U-Net%20architectures%20with%20different%20complexities%20have%20been%20compared%20for%20the%20quantification%20of%20the%20fat%20fraction%20in%20each%20muscle%20group%20selected%20in%20the%20central%20part%20of%20the%20thigh%20region.%20The%20corresponding%20performance%20has%20been%20assessed%20in%20terms%20of%20Dice%20score%20%28DSC%29%20and%20FF%20quantification%20error.%20The%20database%20contained%201450%20thigh%20images%20from%2059%20patients%20and%2014%20healthy%20subjects%20%28age%3A%2047%5Cu2009%5Cu00b1%5Cu200917%5Cu00a0years%2C%20sex%3A%2036F%2C%2037M%29.%20Ten%20individual%20muscles%20were%20segmented%20in%20each%20image.%20The%20performance%20of%20each%20model%20was%20compared%20to%20nnU-Net%2C%20a%20complex%20architecture%20with%204.35%20%5Cu00d7%20107%20parameters%2C%2012.8%20Gigabytes%20of%20peak%20memory%20usage%20and%20167%5Cu00a0h%20of%20training%20time.%5CnRESULTS%3A%20As%20expected%2C%20nnU-Net%20achieved%20the%20highest%20DSC%20%2894.77%5Cu2009%5Cu00b1%5Cu20090.13%25%29.%20A%20simpler%20U-Net%20%285.81%20%5Cu00d7%20105%20parameters%2C%202.37%20Gigabytes%2C%2014%5Cu00a0h%20of%20training%20time%29%20achieved%20a%20lower%20DSC%20but%20still%20above%2090%25.%20Surprisingly%2C%20both%20models%20achieved%20a%20comparable%20FF%20estimation.%5CnDISCUSSION%3A%20The%20poor%20correlation%20between%20observed%20DSC%20and%20FF%20indicates%20that%20less%20complex%20architectures%2C%20reducing%20GPU%20memory%20utilization%20and%20training%20time%2C%20can%20still%20accurately%20quantify%20FF.%22%2C%22date%22%3A%222025-01-11%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1007%5C%2Fs10334-024-01221-3%22%2C%22ISSN%22%3A%221352-8661%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22NW6SZQBR%22%2C%22ZP8JD3DQ%22%2C%22XTA6KS7L%22%5D%2C%22dateModified%22%3A%222025-09-12T07%3A58%3A04Z%22%7D%7D%2C%7B%22key%22%3A%2278WZLEDK%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Nassar%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BNassar%2C%20J.%2C%20Trabelsi%2C%20A.%2C%20Amer%2C%20R.%2C%20Le%20Fur%2C%20Y.%2C%20Attarian%2C%20S.%2C%20Radunsky%2C%20D.%2C%20Blumenfeld-Katzir%2C%20T.%2C%20Greenspan%2C%20H.%2C%20Bendahan%2C%20D.%20and%20Ben-Eliezer%2C%20N.%20%282023%29%20%26%23x201C%3BEstimation%20of%20subvoxel%20fat%20infiltration%20in%20neurodegenerative%20muscle%20disorders%20using%20quantitative%20multi-T2%20analysis%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BNMR%20in%20Biomedicine%26lt%3B%5C%2Fi%26gt%3B%2C%2036%289%29%2C%20p.%20e4947.%20Available%20at%3A%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1002%5C%2Fnbm.4947%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1002%5C%2Fnbm.4947%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Estimation%20of%20subvoxel%20fat%20infiltration%20in%20neurodegenerative%20muscle%20disorders%20using%20quantitative%20multi-T2%20analysis%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jannette%22%2C%22lastName%22%3A%22Nassar%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amira%22%2C%22lastName%22%3A%22Trabelsi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rula%22%2C%22lastName%22%3A%22Amer%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yann%22%2C%22lastName%22%3A%22Le%20Fur%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shahram%22%2C%22lastName%22%3A%22Attarian%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Dvir%22%2C%22lastName%22%3A%22Radunsky%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tamar%22%2C%22lastName%22%3A%22Blumenfeld-Katzir%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Hayit%22%2C%22lastName%22%3A%22Greenspan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22David%22%2C%22lastName%22%3A%22Bendahan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Noam%22%2C%22lastName%22%3A%22Ben-Eliezer%22%7D%5D%2C%22abstractNote%22%3A%22MRI%26%23039%3Bs%20T2%20relaxation%20time%20is%20a%20valuable%20biomarker%20for%20neuromuscular%20disorders%20and%20muscle%20dystrophies.%20One%20of%20the%20hallmarks%20of%20these%20pathologies%20is%20the%20infiltration%20of%20adipose%20tissue%20and%20a%20loss%20of%20muscle%20volume.%20This%20leads%20to%20a%20mixture%20of%20two%20signal%20components%2C%20from%20fat%20and%20from%20water%2C%20to%20appear%20in%20each%20imaged%20voxel%2C%20each%20having%20a%20specific%20T2%20relaxation%20time.%20In%20this%20proof-of-concept%20work%2C%20we%20present%20a%20technique%20that%20can%20separate%20the%20signals%20from%20water%20and%20from%20fat%20within%20each%20voxel%2C%20measure%20their%20separate%20T2%20values%2C%20and%20calculate%20their%20relative%20fractions.%20The%20echo%20modulation%20curve%20%28EMC%29%20algorithm%20is%20a%20dictionary-based%20technique%20that%20offers%20accurate%20and%20reproducible%20mapping%20of%20T2%20relaxation%20times.%20We%20present%20an%20extension%20of%20the%20EMC%20algorithm%20for%20estimating%20subvoxel%20fat%20and%20water%20fractions%2C%20alongside%20the%20T2%20and%20proton-density%20values%20of%20each%20component.%20To%20facilitate%20data%20processing%2C%20calf%20and%20thigh%20anatomy%20were%20automatically%20segmented%20using%20a%20fully%20convolutional%20neural%20network%20and%20FSLeyes%20software.%20The%20preprocessing%20included%20creating%20two%20signal%20dictionaries%2C%20for%20water%20and%20for%20fat%2C%20using%20Bloch%20simulations%20of%20the%20prospective%20protocol.%20Postprocessing%20included%20voxelwise%20fitting%20for%20two%20components%2C%20by%20matching%20the%20experimental%20decay%20curve%20to%20a%20linear%20combination%20of%20the%20two%20simulated%20dictionaries.%20Subvoxel%20fat%20and%20water%20fractions%20and%20relaxation%20times%20were%20generated%20and%20used%20to%20calculate%20a%20new%20quantitative%20biomarker%2C%20termed%20viable%20muscle%20index%2C%20and%20reflecting%20disease%20severity.%20This%20biomarker%20indicates%20the%20fraction%20of%20remaining%20muscle%20out%20of%20the%20entire%20muscle%20region.%20The%20results%20were%20compared%20with%20those%20using%20the%20conventional%20Dixon%20technique%2C%20showing%20high%20agreement%20%28R%20%3D%200.98%2C%20p%20%26lt%3B%200.001%29.%20It%20was%20concluded%20that%20the%20new%20extension%20of%20the%20EMC%20algorithm%20can%20be%20used%20to%20quantify%20abnormal%20fat%20infiltration%20as%20well%20as%20identify%20early%20inflammatory%20processes%20corresponding%20to%20elevation%20in%20the%20T2%20value%20of%20the%20water%20%28muscle%29%20component.%20This%20new%20ability%20may%20improve%20the%20diagnostic%20accuracy%20of%20neuromuscular%20diseases%2C%20help%20stratification%20of%20patients%20according%20to%20disease%20severity%2C%20and%20offer%20an%20efficient%20tool%20for%20tracking%20disease%20progression.%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1002%5C%2Fnbm.4947%22%2C%22ISSN%22%3A%221099-1492%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1002%5C%2Fnbm.4947%22%2C%22collections%22%3A%5B%22SVI8UJ6H%22%2C%22NW6SZQBR%22%5D%2C%22dateModified%22%3A%222025-09-11T07%3A32%3A30Z%22%7D%7D%2C%7B%22key%22%3A%2237VZFDGX%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Durelle%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BDurelle%2C%20C.%2C%20Delmont%2C%20E.%2C%20Michel%2C%20C.%2C%20Trabelsi%2C%20A.%2C%20Hostin%2C%20M.-A.%2C%20Ogier%2C%20A.%2C%20Bendahan%2C%20D.%20and%20Attarian%2C%20S.%20%282023%29%20%26%23x201C%3BQuantification%20of%20muscle%20involvement%20in%20familial%20amyloid%20polyneuropathy%20using%20MRI%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BEuropean%20Journal%20of%20Neurology%26lt%3B%5C%2Fi%26gt%3B%2C%2030%2810%29%2C%20pp.%203286%26%23x2013%3B3295.%20Available%20at%3A%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1111%5C%2Fene.15970%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1111%5C%2Fene.15970%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Quantification%20of%20muscle%20involvement%20in%20familial%20amyloid%20polyneuropathy%20using%20MRI%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Cl%5Cu00e9mence%22%2C%22lastName%22%3A%22Durelle%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Emilien%22%2C%22lastName%22%3A%22Delmont%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Constance%22%2C%22lastName%22%3A%22Michel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amira%22%2C%22lastName%22%3A%22Trabelsi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marc-Adrien%22%2C%22lastName%22%3A%22Hostin%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Augustin%22%2C%22lastName%22%3A%22Ogier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22David%22%2C%22lastName%22%3A%22Bendahan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shahram%22%2C%22lastName%22%3A%22Attarian%22%7D%5D%2C%22abstractNote%22%3A%22Background%20and%20purpose%20Transthyretin%20familial%20amyloid%20polyneuropathy%20%28TTR-FAP%29%20is%20a%20rare%20genetic%20disease%20with%20autosomal-dominant%20inheritance.%20In%20this%20study%2C%20we%20aimed%20to%20quantify%20fatty%20infiltration%20%28fat%20fraction%20%5BFF%5D%29%20and%20magnetization%20transfer%20ratio%20%28MTR%29%20in%20individual%20muscles%20of%20patients%20with%20symptomatic%20and%20asymptomatic%20TTR-FAP%20using%20magnetic%20resonance%20imaging.%20Secondarily%2C%20we%20aimed%20to%20assess%20correlations%20with%20clinical%20and%20electrophysiological%20variables.%20Methods%20A%20total%20of%2039%20patients%20with%20a%20confirmed%20mutation%20in%20the%20TTR%20gene%20%2825%20symptomatic%20and%2014%20asymptomatic%29%20and%2014%20healthy%20volunteers%20were%20included.%20A%20total%20of%2016%20muscles%20were%20manually%20delineated%20in%20the%20nondominant%20lower%20limb%20from%20T1-weighted%20anatomical%20images.%20The%20corresponding%20masks%20were%20propagated%20on%20the%20MTR%20and%20FF%20maps.%20Detailed%20neurological%20and%20electrophysiological%20examinations%20were%20conducted%20in%20each%20group.%20Results%20The%20MTR%20was%20decreased%20%2842.6%20AU%3B%20p%20%3D%200.001%29%20and%20FF%20was%20elevated%20%2814%25%3B%20p%20%3D%200.003%29%20in%20the%20lower%20limbs%20of%20the%20symptomatic%20group%2C%20with%20preferential%20posterior%20and%20lateral%20involvement.%20In%20the%20asymptomatic%20group%2C%20elevated%20FF%20was%20quantified%20in%20the%20gastrocnemius%20lateralis%20muscle%20%2811%25%3B%20p%20%3D%200.021%29.%20FF%20was%20significantly%20correlated%20with%20disease%20duration%20%28r%20%3D%200.49%2C%20p%20%3D%200.015%29%2C%20neuropathy%20impairment%20score%20for%20the%20lower%20limb%20%28r%20%3D%200.42%2C%20p%20%3D%200.041%29%2C%20Overall%20Neuropathy%20Limitations%20Scale%20score%20%28r%20%3D%200.49%2C%20p%20%3D%200.013%29%2C%20polyneuropathy%20disability%20score%20%28r%20%3D%200.57%2C%20p%20%3D%200.03%29%20and%20the%20sum%20of%20compound%20muscle%20action%20potential%20%28r%20%3D%200.52%2C%20p%20%3D%200.009%29.%20MTR%20was%20strongly%20correlated%20to%20FF%20%28r%20%3D%200.78%2C%20p%20%26lt%3B%200.0001%29%2C%20and%20a%20few%20muscles%20with%20an%20FF%20within%20the%20normal%20range%20had%20a%20reduced%20MTR.%20Conclusion%20These%20observations%20suggest%20that%20FF%20and%20MTR%20could%20be%20interesting%20biomarkers%20in%20TTR-FAP.%20In%20asymptomatic%20patients%2C%20FF%20in%20the%20gastrocnemius%20lateralis%20muscle%20could%20be%20a%20good%20indicator%20of%20the%20transition%20from%20an%20asymptomatic%20to%20a%20symptomatic%20form%20of%20the%20disease.%20MTR%20could%20be%20an%20early%20biomarker%20of%20muscle%20alterations.%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.1111%5C%2Fene.15970%22%2C%22ISSN%22%3A%221468-1331%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fonlinelibrary.wiley.com%5C%2Fdoi%5C%2Fabs%5C%2F10.1111%5C%2Fene.15970%22%2C%22collections%22%3A%5B%22NW6SZQBR%22%2C%2248SZ2I24%22%5D%2C%22dateModified%22%3A%222025-09-11T07%3A32%3A30Z%22%7D%7D%2C%7B%22key%22%3A%22CUDJ6CSH%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Amer%20et%20al.%22%2C%22parsedDate%22%3A%222022-07-14%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BAmer%2C%20R.%2C%20Nassar%2C%20J.%2C%20Trabelsi%2C%20A.%2C%20Bendahan%2C%20D.%2C%20Greenspan%2C%20H.%20and%20Ben-Eliezer%2C%20N.%20%282022%29%20%26%23x201C%3BQuantification%20of%20Intra-Muscular%20Adipose%20Infiltration%20in%20Calf%5C%2FThigh%20MRI%20Using%20Fully%20and%20Weakly%20Supervised%20Semantic%20Segmentation%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BBioengineering%26lt%3B%5C%2Fi%26gt%3B%2C%209%287%29%2C%20p.%20315.%20Available%20at%3A%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.3390%5C%2Fbioengineering9070315%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.3390%5C%2Fbioengineering9070315%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Quantification%20of%20Intra-Muscular%20Adipose%20Infiltration%20in%20Calf%5C%2FThigh%20MRI%20Using%20Fully%20and%20Weakly%20Supervised%20Semantic%20Segmentation%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rula%22%2C%22lastName%22%3A%22Amer%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jannette%22%2C%22lastName%22%3A%22Nassar%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Amira%22%2C%22lastName%22%3A%22Trabelsi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22David%22%2C%22lastName%22%3A%22Bendahan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Hayit%22%2C%22lastName%22%3A%22Greenspan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Noam%22%2C%22lastName%22%3A%22Ben-Eliezer%22%7D%5D%2C%22abstractNote%22%3A%22Purpose%3A%20Infiltration%20of%20fat%20into%20lower%20limb%20muscles%20is%20one%20of%20the%20key%20markers%20for%20the%20severity%20of%20muscle%20pathologies.%20The%20level%20of%20fat%20infiltration%20varies%20in%20its%20severity%20across%20and%20within%20patients%2C%20and%20it%20is%20traditionally%20estimated%20using%20visual%20radiologic%20inspection.%20Precise%20quantification%20of%20the%20severity%20and%20spatial%20distribution%20of%20this%20pathological%20process%20requires%20accurate%20segmentation%20of%20lower%20limb%20anatomy%20into%20muscle%20and%20fat.%20Methods%3A%20Quantitative%20magnetic%20resonance%20imaging%20%28qMRI%29%20of%20the%20calf%20and%20thigh%20muscles%20is%20one%20of%20the%20most%20effective%20techniques%20for%20estimating%20pathological%20accumulation%20of%20intra-muscular%20adipose%20tissue%20%28IMAT%29%20in%20muscular%20dystrophies.%20In%20this%20work%2C%20we%20present%20a%20new%20deep%20learning%20%28DL%29%20network%20tool%20for%20automated%20and%20robust%20segmentation%20of%20lower%20limb%20anatomy%20that%20is%20based%20on%20the%20quantification%20of%20MRI%5Cu2019s%20transverse%20%28T2%29%20relaxation%20time.%20The%20network%20was%20used%20to%20segment%20calf%20and%20thigh%20anatomies%20into%20viable%20muscle%20areas%20and%20IMAT%20using%20a%20weakly%20supervised%20learning%20process.%20A%20new%20disease%20biomarker%20was%20calculated%2C%20reflecting%20the%20level%20of%20abnormal%20fat%20infiltration%20and%20disease%20state.%20A%20biomarker%20was%20then%20applied%20on%20two%20patient%20populations%20suffering%20from%20dysferlinopathy%20and%20Charcot%5Cu2013Marie%5Cu2013Tooth%20%28CMT%29%20diseases.%20Results%3A%20Comparison%20of%20manual%20vs.%20automated%20segmentation%20of%20muscle%20anatomy%2C%20viable%20muscle%20areas%2C%20and%20intermuscular%20adipose%20tissue%20%28IMAT%29%20produced%20high%20Dice%20similarity%20coefficients%20%28DSCs%29%20of%2096.4%25%2C%2091.7%25%2C%20and%2093.3%25%2C%20respectively.%20Linear%20regression%20between%20the%20biomarker%20value%20calculated%20based%20on%20the%20ground%20truth%20segmentation%20and%20based%20on%20automatic%20segmentation%20produced%20high%20correlation%20coefficients%20of%2097.7%25%20and%2095.9%25%20for%20the%20dysferlinopathy%20and%20CMT%20patients%2C%20respectively.%20Conclusions%3A%20Using%20a%20combination%20of%20qMRI%20and%20DL-based%20segmentation%2C%20we%20present%20a%20new%20quantitative%20biomarker%20of%20disease%20severity.%20This%20biomarker%20is%20automatically%20calculated%20and%2C%20most%20importantly%2C%20provides%20a%20spatially%20global%20indication%20for%20the%20state%20of%20the%20disease%20across%20the%20entire%20thigh%20or%20calf.%22%2C%22date%22%3A%222022-07-14%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.3390%5C%2Fbioengineering9070315%22%2C%22ISSN%22%3A%222306-5354%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.mdpi.com%5C%2F2306-5354%5C%2F9%5C%2F7%5C%2F315%22%2C%22collections%22%3A%5B%22NW6SZQBR%22%5D%2C%22dateModified%22%3A%222025-09-11T07%3A32%3A31Z%22%7D%7D%2C%7B%22key%22%3A%22KZVPYUJD%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22lastModifiedByUser%22%3A%7B%22id%22%3A1497928%2C%22username%22%3A%22fkober%22%2C%22name%22%3A%22Frank%20Kober%22%2C%22links%22%3A%7B%22alternate%22%3A%7B%22href%22%3A%22https%3A%5C%2F%5C%2Fwww.zotero.org%5C%2Ffkober%22%2C%22type%22%3A%22text%5C%2Fhtml%22%7D%7D%7D%2C%22creatorSummary%22%3A%22Rasoanandrianina%20et%20al.%22%2C%22parsedDate%22%3A%222020-05-01%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BRasoanandrianina%2C%20H.%2C%20Demorti%26%23xE8%3Bre%2C%20S.%2C%20Trabelsi%2C%20A.%2C%20Ranjeva%2C%20J.P.%2C%20Girard%2C%20O.%2C%20Duhamel%2C%20G.%2C%20Guye%2C%20M.%2C%20Pelletier%2C%20J.%2C%20Audoin%2C%20B.%20and%20Callot%2C%20V.%20%282020%29%20%26%23x201C%3BSensitivity%20of%20the%20Inhomogeneous%20Magnetization%20Transfer%20Imaging%20Technique%20to%20Spinal%20Cord%20Damage%20in%20Multiple%20Sclerosis%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BAmerican%20Journal%20of%20Neuroradiology%26lt%3B%5C%2Fi%26gt%3B%2C%2041%285%29%2C%20pp.%20929%26%23x2013%3B937.%20Available%20at%3A%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.3174%5C%2Fajnr.A6554%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.3174%5C%2Fajnr.A6554%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Sensitivity%20of%20the%20Inhomogeneous%20Magnetization%20Transfer%20Imaging%20Technique%20to%20Spinal%20Cord%20Damage%20in%20Multiple%20Sclerosis%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22H.%22%2C%22lastName%22%3A%22Rasoanandrianina%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22S.%22%2C%22lastName%22%3A%22Demorti%5Cu00e8re%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22A.%22%2C%22lastName%22%3A%22Trabelsi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22J.%20P.%22%2C%22lastName%22%3A%22Ranjeva%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22O.%22%2C%22lastName%22%3A%22Girard%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22G.%22%2C%22lastName%22%3A%22Duhamel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22M.%22%2C%22lastName%22%3A%22Guye%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22J.%22%2C%22lastName%22%3A%22Pelletier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22B.%22%2C%22lastName%22%3A%22Audoin%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22V.%22%2C%22lastName%22%3A%22Callot%22%7D%5D%2C%22abstractNote%22%3A%22BACKGROUND%20AND%20PURPOSE%3A%20The%20inhomogeneous%20magnetization%20transfer%20technique%20has%20demonstrated%20high%20specificity%20for%20myelin%2C%20and%20has%20shown%20sensitivity%20to%20multiple%20sclerosis-related%20impairment%20in%20brain%20tissue.%20Our%20aim%20was%20to%20investigate%20its%20sensitivity%20to%20spinal%20cord%20impairment%20in%20MS%20relative%20to%20more%20established%20MR%20imaging%20techniques%20%28volumetry%2C%20magnetization%20transfer%2C%20DTI%29.%5CnMATERIALS%20AND%20METHODS%3A%20Anatomic%20images%20covering%20the%20cervical%20spinal%20cord%20from%20the%20C1%20to%20C6%20levels%20and%20DTI%2C%20magnetization%20transfer%5C%2Finhomogeneous%20magnetization%20transfer%20images%20at%20the%20C2%5C%2FC5%20levels%20were%20acquired%20in%2019%20patients%20with%20MS%20and%2019%20paired%20healthy%20controls.%20Anatomic%20images%20were%20segmented%20in%20spinal%20cord%20GM%20and%20WM%2C%20both%20manually%20and%20using%20the%20AMU40%20atlases.%20MS%20lesions%20were%20manually%20delineated.%20MR%20metrics%20were%20analyzed%20within%20normal-appearing%20and%20lesion%20regions%20in%20anterolateral%20and%20posterolateral%20WM%20and%20compared%20using%20Wilcoxon%20rank%20tests%20and%20z%20scores.%20Correlations%20between%20MR%20metrics%20and%20clinical%20scores%20in%20patients%20with%20MS%20were%20evaluated%20using%20the%20Spearman%20rank%20correlation.%5CnRESULTS%3A%20AMU40-based%20C1-to-C6%20GM%5C%2FWM%20automatic%20segmentations%20in%20patients%20with%20MS%20were%20evaluated%20relative%20to%20manual%20delineation.%20Mean%20Dice%20coefficients%20were%200.75%5C%2F0.89%2C%20respectively.%20All%20MR%20metrics%20%28WM%5C%2FGM%20cross-sectional%20areas%2C%20normal-appearing%20and%20lesion%20diffusivities%2C%20and%20magnetization%20transfer%5C%2Finhomogeneous%20magnetization%20transfer%20ratios%29%20were%20observed%20altered%20in%20patients%20compared%20with%20controls%20%28P%20%26lt%3B%20.05%29.%20Additionally%2C%20the%20absolute%20inhomogeneous%20magnetization%20transfer%20ratio%20z%20scores%20were%20significantly%20higher%20than%20those%20of%20the%20other%20MR%20metrics%20%28P%20%26lt%3B%20.0001%29%2C%20suggesting%20a%20higher%20inhomogeneous%20magnetization%20transfer%20sensitivity%20toward%20spinal%20cord%20impairment%20in%20MS.%20Significant%20correlations%20with%20the%20Expanded%20Disability%20Status%20Scale%20%28%5Cu03c1%20%3D%20%5Cu20130.73%5C%2FP%20%3D%20.02%2C%20%5Cu03c1%20%3D%20%5Cu20130.81%5C%2FP%20%3D%20.004%29%20and%20the%20total%20Medical%20Research%20Council%20scale%20%28%5Cu03c1%20%3D%200.80%5C%2FP%20%3D%20.009%2C%20%5Cu03c1%20%3D%20%5Cu20130.74%5C%2FP%20%3D%20.02%29%20were%20observed%20for%20inhomogeneous%20magnetization%20transfer%20and%20magnetization%20transfer%20ratio%20z%20scores%2C%20respectively%2C%20in%20normal-appearing%20WM%20regions%2C%20while%20weaker%20and%20nonsignificant%20correlations%20were%20obtained%20for%20DTI%20metrics.%5CnCONCLUSIONS%3A%20With%20inhomogeneous%20magnetization%20transfer%20being%20highly%20sensitive%20to%20spinal%20cord%20damage%20in%20MS%20compared%20with%20conventional%20magnetization%20transfer%20and%20DTI%2C%20it%20could%20generate%20great%20clinical%20interest%20for%20longitudinal%20follow-up%20and%20potential%20remyelinating%20clinical%20trials.%20In%20line%20with%20other%20advanced%20myelin%20techniques%20with%20which%20it%20could%20be%20compared%2C%20it%20opens%20perspectives%20for%20multicentric%20investigations.%22%2C%22date%22%3A%222020%5C%2F05%5C%2F01%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.3174%5C%2Fajnr.A6554%22%2C%22ISSN%22%3A%220195-6108%2C%201936-959X%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Fwww.ajnr.org%5C%2Fcontent%5C%2F41%5C%2F5%5C%2F929%22%2C%22collections%22%3A%5B%2274RX9758%22%2C%22ZFC2Q648%22%2C%22UP3H3ICX%22%2C%22XTA6KS7L%22%5D%2C%22dateModified%22%3A%222025-09-11T07%3A32%3A29Z%22%7D%7D%5D%7D
Martin, S., Trabelsi, A., Guye, M., Dubois, M., Abdeddaim, R., Bendahan, D. and André, R. (2025) “Optimized reconstruction of undersampled Dixon sequences using new memory-efficient unrolled deep neural networks: HalfVarNet and HalfDIRCN,” Magnetic Resonance in Medicine [Preprint]. Available at: https://doi.org/10.1002/mrm.70070.
Martin, S., André, R., Trabelsi, A., Michel, C.P., Fortanier, E., Attarian, S., Guye, M., Dubois, M., Abdeddaim, R. and Bendahan, D. (2025) “Importance of neural network complexity for the automatic segmentation of individual thigh muscles in MRI images from patients with neuromuscular diseases,” Magma (New York, N.Y.) [Preprint]. Available at: https://doi.org/10.1007/s10334-024-01221-3.
Nassar, J., Trabelsi, A., Amer, R., Le Fur, Y., Attarian, S., Radunsky, D., Blumenfeld-Katzir, T., Greenspan, H., Bendahan, D. and Ben-Eliezer, N. (2023) “Estimation of subvoxel fat infiltration in neurodegenerative muscle disorders using quantitative multi-T2 analysis,” NMR in Biomedicine, 36(9), p. e4947. Available at: https://doi.org/10.1002/nbm.4947.
Durelle, C., Delmont, E., Michel, C., Trabelsi, A., Hostin, M.-A., Ogier, A., Bendahan, D. and Attarian, S. (2023) “Quantification of muscle involvement in familial amyloid polyneuropathy using MRI,” European Journal of Neurology, 30(10), pp. 3286–3295. Available at: https://doi.org/10.1111/ene.15970.
Amer, R., Nassar, J., Trabelsi, A., Bendahan, D., Greenspan, H. and Ben-Eliezer, N. (2022) “Quantification of Intra-Muscular Adipose Infiltration in Calf/Thigh MRI Using Fully and Weakly Supervised Semantic Segmentation,” Bioengineering, 9(7), p. 315. Available at: https://doi.org/10.3390/bioengineering9070315.
Rasoanandrianina, H., Demortière, S., Trabelsi, A., Ranjeva, J.P., Girard, O., Duhamel, G., Guye, M., Pelletier, J., Audoin, B. and Callot, V. (2020) “Sensitivity of the Inhomogeneous Magnetization Transfer Imaging Technique to Spinal Cord Damage in Multiple Sclerosis,” American Journal of Neuroradiology, 41(5), pp. 929–937. Available at: https://doi.org/10.3174/ajnr.A6554.