Fournel Joris
Publications :
180164
Fournel
1
harvard-cite-them-right-no-et-al
50
date
desc
year
4818
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%22UXT6JCKV%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Fournel%20et%20al.%22%2C%22parsedDate%22%3A%222025-11-07%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%3BFournel%2C%20J.%2C%20De%20Masi%2C%20M.%2C%20Lu%2C%20C.%2C%20Omnes%2C%20V.%2C%20Muselier%2C%20B.%2C%20Ghattas%2C%20B.%2C%20Bouchot%2C%20O.%2C%20Kafi%2C%20M.%2C%20Lalande%2C%20A.%2C%20Gaudry%2C%20M.%2C%20Jacquier%2C%20A.%20and%20Bartoli%2C%20A.%20%282025%29%20%26%23x201C%3BDeep%20learning-driven%20false-lumen%20volumes%20predict%20adverse%20remodeling%20better%20than%20diameter%20in%20patients%20with%20residual%20aortic%20dissection%20on%20CT%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BEuropean%20Radiology%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%2Fs00330-025-12116-9%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1007%5C%2Fs00330-025-12116-9%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%22Deep%20learning-driven%20false-lumen%20volumes%20predict%20adverse%20remodeling%20better%20than%20diameter%20in%20patients%20with%20residual%20aortic%20dissection%20on%20CT%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Joris%22%2C%22lastName%22%3A%22Fournel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Mariangela%22%2C%22lastName%22%3A%22De%20Masi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Charlotte%22%2C%22lastName%22%3A%22Lu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Virgile%22%2C%22lastName%22%3A%22Omnes%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Baptiste%22%2C%22lastName%22%3A%22Muselier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Badih%22%2C%22lastName%22%3A%22Ghattas%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Olivier%22%2C%22lastName%22%3A%22Bouchot%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Moundji%22%2C%22lastName%22%3A%22Kafi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alain%22%2C%22lastName%22%3A%22Lalande%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marine%22%2C%22lastName%22%3A%22Gaudry%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Jacquier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Axel%22%2C%22lastName%22%3A%22Bartoli%22%7D%5D%2C%22abstractNote%22%3A%22OBJECTIVES%3A%201.%20To%20develop%20a%20deep-learning%20segmentation%20model%20for%20automated%20measurement%20of%20maximal%20aortic%20diameter%20%28Dmax%29%20and%20volumes%20of%20aortic%20dissection%20components%3A%20true-lumen%20%28TL%29%2C%20circulating%20false-lumen%20%28CFL%29%2C%20and%20thrombus%20%28Th%29%20on%20CT%20angiography%20%28CTA%29.%202.%20To%20assess%20the%20predictive%20value%20of%20these%20measures%20for%20adverse%20aortic%20remodeling%20in%20residual%20aortic%20dissection%20%28RAD%29.%5CnMATERIALS%20AND%20METHODS%3A%20This%20retrospective%20study%20included%20322%20patients%20from%20two%20centers.%20The%20segmentation%20model%20was%20trained%20on%20120%20patients%20%28Center%201%29%20and%20tested%20on%20an%20internal%20dataset%20%2830%20patients%2C%20Center%201%29%20and%20an%20external%20dataset%20%2810%20patients%2C%20Center%202%29%20in%20terms%20of%20Dice%20Similarity%20Coefficient%20%28DSC%29.%20The%20model%20extracted%20Dmax%2C%20global%20false-lumen%20volume%20%28FLGlo%5Cu2009%3D%5Cu2009CFL%5Cu2009%2B%5Cu2009Th%29%2C%20and%20local%20false-lumen%20volume%20%28FLLoc%2C%20measured%203%5Cu2009cm%20around%20the%20largest%20diameter%29.%20Clinical%20validation%20was%20performed%20on%2083%20patients%20from%20Center1%20%28internal%20validation%2C%202-year%20follow-up%29%20and%2079%20patients%20from%20Center2%20%28external%20validation%2C%204.5-year%20follow-up%29.%5CnRESULTS%3A%20The%20segmentation%20model%20achieved%20high%20accuracy%20%28Center%201%2C%20DSC%3A%200.93%20TL%2C%200.93%20CFL%2C%200.87%20Th%3B%20Center%202%2C%20DSC%3A%200.92%20TL%2C%200.93%20CFL%2C%200.84%20Th%29%20with%20strong%20agreement%20between%20automated%20and%20manual%20measurements.%20Aortic%20remodeling%20occurred%20in%2039%5C%2F83%20patients%20%2846.9%25%29%20from%20Center1%20and%2033%5C%2F79%20patients%20%2841.7%25%29%20from%20Center2.%20Aortic%20remodeling%20occurred%20in%2039%5C%2F83%20patients%20%2847%25%29%20from%20Center1%20and%2033%5C%2F80%20%2842%25%29%20from%20Center2.%20FLLoc%20outperformed%20Dmax%20and%20FLGlo%20%28Center%201%3A%20AUC%5Cu2009%3D%5Cu20090.83%2C%200.73%2C%20and%200.76%3B%20Center%202%3A%20AUC%5Cu2009%3D%5Cu20090.77%2C%200.64%2C%20and%200.70%29.%20At%20optimal%20thresholds%2C%20FLLoc%20showed%20good%20predictive%20performance%20%28Center%201%3A%20Sensitivity%5Cu2009%3D%5Cu20090.87%2C%20Specificity%5Cu2009%3D%5Cu20090.68%29.%5CnCONCLUSION%3A%20Deep-learning%20segmentation%20provides%20accurate%20aortic%20measurements.%20Local%20false-lumen%20volumes%20predict%20adverse%20aortic%20remodeling%20in%20RAD%20better%20than%20diameter%20and%20global%20false-lumen%20volumes.%5CnKEY%20POINTS%3A%20Question%20In%20residual%20aortic%20dissection%20%28RAD%29%20after%20type-A%20dissection%2C%20early%20identification%20of%20high-risk%20patients%20on%20initial%20CT%20angiography%20is%20crucial%20for%20endovascular%20treatment%20decisions.%20Findings%20False-lumen%20local%20volumes%20%283%5Cu2009cm%20around%20aortic%20dissection%20maximal%20diameters%29%2C%20obtained%20with%20an%20automatic%20deep-learning%20method%2C%20predict%20adverse%20remodeling%20better%20than%20diameter%20or%20global%20false-lumen%20volumes.%20Clinical%20relevance%20A%20deep-learning%20segmentation%20method%20of%20aortic%20dissection%20components%20on%20CTA%2C%20enabling%20automatic%20measurements%20of%20diameters%20and%20volumes%20is%20feasible.%20It%20provides%20local%20false-lumen%20volumes%2C%20a%20better%20predictive%20marker%20of%20adverse%20aortic%20remodeling%20than%20the%20currently%20used%20diameters%20and%20global%20volumes.%22%2C%22date%22%3A%222025-11-07%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1007%5C%2Fs00330-025-12116-9%22%2C%22ISSN%22%3A%221432-1084%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22KFZ6H27M%22%2C%22CXAMISN8%22%5D%2C%22dateModified%22%3A%222025-11-27T10%3A53%3A30Z%22%7D%7D%2C%7B%22key%22%3A%22AK7H9LVJ%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Gaborit%20et%20al.%22%2C%22parsedDate%22%3A%222024-09-03%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%3BGaborit%2C%20B.%2C%20Julla%2C%20J.B.%2C%20Fournel%2C%20J.%2C%20Ancel%2C%20P.%2C%20Soghomonian%2C%20A.%2C%20Deprade%2C%20C.%2C%20Lasbleiz%2C%20A.%2C%20Houssays%2C%20M.%2C%20Ghattas%2C%20B.%2C%20Gascon%2C%20P.%2C%20Righini%2C%20M.%2C%20Matonti%2C%20F.%2C%20Venteclef%2C%20N.%2C%20Potier%2C%20L.%2C%20Gautier%2C%20J.F.%2C%20Resseguier%2C%20N.%2C%20Bartoli%2C%20A.%2C%20Mourre%2C%20F.%2C%20Darmon%2C%20P.%2C%20Jacquier%2C%20A.%20and%20Dutour%2C%20A.%20%282024%29%20%26%23x201C%3BFully%20automated%20epicardial%20adipose%20tissue%20volume%20quantification%20with%20deep%20learning%20and%20relationship%20with%20CAC%20score%20and%20micro%5C%2Fmacrovascular%20complications%20in%20people%20living%20with%20type%202%20diabetes%3A%20the%20multicenter%20EPIDIAB%20study%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BCardiovascular%20Diabetology%26lt%3B%5C%2Fi%26gt%3B%2C%2023%281%29%2C%20p.%20328.%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%2F10.1186%5C%2Fs12933-024-02411-y%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1186%5C%2Fs12933-024-02411-y%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%22Fully%20automated%20epicardial%20adipose%20tissue%20volume%20quantification%20with%20deep%20learning%20and%20relationship%20with%20CAC%20score%20and%20micro%5C%2Fmacrovascular%20complications%20in%20people%20living%20with%20type%202%20diabetes%3A%20the%20multicenter%20EPIDIAB%20study%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22B%5Cu00e9n%5Cu00e9dicte%22%2C%22lastName%22%3A%22Gaborit%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jean%20Baptiste%22%2C%22lastName%22%3A%22Julla%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Joris%22%2C%22lastName%22%3A%22Fournel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Patricia%22%2C%22lastName%22%3A%22Ancel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Astrid%22%2C%22lastName%22%3A%22Soghomonian%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Camille%22%2C%22lastName%22%3A%22Deprade%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ad%5Cu00e8le%22%2C%22lastName%22%3A%22Lasbleiz%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Marie%22%2C%22lastName%22%3A%22Houssays%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Badih%22%2C%22lastName%22%3A%22Ghattas%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Pierre%22%2C%22lastName%22%3A%22Gascon%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Maud%22%2C%22lastName%22%3A%22Righini%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fr%5Cu00e9d%5Cu00e9ric%22%2C%22lastName%22%3A%22Matonti%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Nicolas%22%2C%22lastName%22%3A%22Venteclef%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Louis%22%2C%22lastName%22%3A%22Potier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jean%20Fran%5Cu00e7ois%22%2C%22lastName%22%3A%22Gautier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22No%5Cu00e9mie%22%2C%22lastName%22%3A%22Resseguier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Axel%22%2C%22lastName%22%3A%22Bartoli%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Florian%22%2C%22lastName%22%3A%22Mourre%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Patrice%22%2C%22lastName%22%3A%22Darmon%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Jacquier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Anne%22%2C%22lastName%22%3A%22Dutour%22%7D%5D%2C%22abstractNote%22%3A%22BACKGROUND%3A%20The%20aim%20of%20this%20study%20%28EPIDIAB%29%20was%20to%20assess%20the%20relationship%20between%20epicardial%20adipose%20tissue%20%28EAT%29%20and%20the%20micro%20and%20macrovascular%20complications%20%28MVC%29%20of%20type%202%20diabetes%20%28T2D%29.%5CnMETHODS%3A%20EPIDIAB%20is%20a%20post%20hoc%20analysis%20from%20the%20AngioSafe%20T2D%20study%2C%20which%20is%20a%20multicentric%20study%20aimed%20at%20determining%20the%20safety%20of%20antihyperglycemic%20drugs%20on%20retina%20and%20including%20patients%20with%20T2D%20screened%20for%20diabetic%20retinopathy%20%28DR%29%20%28n%5Cu2009%3D%5Cu20097200%29%20and%20deeply%20phenotyped%20for%20MVC.%20Patients%20included%20who%20had%20undergone%20cardiac%20CT%20for%20CAC%20%28Coronary%20Artery%20Calcium%29%20scoring%20after%20inclusion%20%28n%5Cu2009%3D%5Cu20091253%29%20were%20tested%20with%20a%20validated%20deep%20learning%20segmentation%20pipeline%20for%20EAT%20volume%20quantification.%5CnRESULTS%3A%20Median%20age%20of%20the%20study%20population%20was%2061%20%5B54%3B67%5D%2C%20with%20a%20majority%20of%20men%20%2857%25%29%20a%20median%20duration%20of%20the%20disease%2011%5Cu00a0years%20%5B5%3B18%5D%20and%20a%20mean%20HbA1c%20of7.8%5Cu2009%5Cu00b1%5Cu20091.4%25.%20EAT%20was%20significantly%20associated%20with%20all%20traditional%20CV%20risk%20factors.%20EAT%20volume%20significantly%20increased%20with%20chronic%20kidney%20disease%20%28CKD%20vs%20no%20CKD%3A%2087.8%20%5B63.5%3B118.6%5D%20vs%2082.7%5Cu00a0mL%20%5B58.8%3B110.8%5D%2C%20p%5Cu2009%3D%5Cu20090.008%29%2C%20coronary%20artery%20disease%20%28CAD%20vs%20no%20CAD%3A%20112.2%20%5B82.7%3B133.3%5D%20vs%2083.8%5Cu00a0mL%20%5B59.4%3B112.1%5D%2C%20p%5Cu2009%3D%5Cu20090.0004%2C%20peripheral%20arterial%20disease%20%28PAD%20vs%20no%20PAD%3A%20107%20%5B76.2%3B141%5D%20vs%2084.6%5Cu00a0mL%5B59.2%3B%20114%5D%2C%20p%5Cu2009%3D%5Cu20090.0005%20and%20elevated%20CAC%20score%20%28%26gt%3B%5Cu2009100%20vs%20%5Cu2009%26lt%3B%5Cu2009100%20AU%3A%2096.8%5Cu00a0mL%20%5B69.1%3B130%5D%20vs%2077.9%5Cu00a0mL%20%5B53.8%3B107.7%5D%2C%20p%5Cu2009%26lt%3B%5Cu20090.0001%29.%20By%20contrast%2C%20EAT%20volume%20was%20neither%20associated%20with%20DR%2C%20nor%20with%20peripheral%20neuropathy.%20We%20further%20evidenced%20a%20subgroup%20of%20patients%20with%20high%20EAT%20volume%20and%20a%20null%20CAC%20score.%20Interestingly%2C%20this%20group%20were%20more%20likely%20to%20be%20composed%20of%20young%20women%20with%20a%20high%20BMI%2C%20a%20lower%20duration%20of%20T2D%2C%20a%20lower%20prevalence%20of%20microvascular%20complications%2C%20and%20a%20higher%20inflammatory%20profile.%5CnCONCLUSIONS%3A%20Fully-automated%20EAT%20volume%20quantification%20could%20provide%20useful%20information%20about%20the%20risk%20of%20both%20renal%20and%20macrovascular%20complications%20in%20T2D%20patients.%22%2C%22date%22%3A%222024-09-03%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1186%5C%2Fs12933-024-02411-y%22%2C%22ISSN%22%3A%221475-2840%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22KFZ6H27M%22%5D%2C%22dateModified%22%3A%222025-11-27T10%3A53%3A31Z%22%7D%7D%2C%7B%22key%22%3A%22D3QFCXE7%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Aung%20et%20al.%22%2C%22parsedDate%22%3A%222024-04%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%3BAung%2C%20N.%2C%20Bartoli%2C%20A.%2C%20Rauseo%2C%20E.%2C%20Cortaredona%2C%20S.%2C%20Sanghvi%2C%20M.M.%2C%20Fournel%2C%20J.%2C%20Ghattas%2C%20B.%2C%20Khanji%2C%20M.Y.%2C%20Petersen%2C%20S.E.%20and%20Jacquier%2C%20A.%20%282024%29%20%26%23x201C%3BLeft%20Ventricular%20Trabeculations%20at%20Cardiac%20MRI%3A%20Reference%20Ranges%20and%20Association%20with%20Cardiovascular%20Risk%20Factors%20in%20UK%20Biobank%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BRadiology%26lt%3B%5C%2Fi%26gt%3B%2C%20311%281%29%2C%20p.%20e232455.%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%2F10.1148%5C%2Fradiol.232455%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1148%5C%2Fradiol.232455%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%22Left%20Ventricular%20Trabeculations%20at%20Cardiac%20MRI%3A%20Reference%20Ranges%20and%20Association%20with%20Cardiovascular%20Risk%20Factors%20in%20UK%20Biobank%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Nay%22%2C%22lastName%22%3A%22Aung%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Axel%22%2C%22lastName%22%3A%22Bartoli%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Elisa%22%2C%22lastName%22%3A%22Rauseo%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sebastien%22%2C%22lastName%22%3A%22Cortaredona%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Mihir%20M.%22%2C%22lastName%22%3A%22Sanghvi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Joris%22%2C%22lastName%22%3A%22Fournel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Badih%22%2C%22lastName%22%3A%22Ghattas%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Mohammed%20Y.%22%2C%22lastName%22%3A%22Khanji%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Steffen%20E.%22%2C%22lastName%22%3A%22Petersen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Jacquier%22%7D%5D%2C%22abstractNote%22%3A%22Background%20The%20extent%20of%20left%20ventricular%20%28LV%29%20trabeculation%20and%20its%20relationship%20with%20cardiovascular%20%28CV%29%20risk%20factors%20is%20unclear.%20Purpose%20To%20apply%20automated%20segmentation%20to%20UK%20Biobank%20cardiac%20MRI%20scans%20to%20%28a%29%20assess%20the%20association%20between%20individual%20characteristics%20and%20CV%20risk%20factors%20and%20trabeculated%20LV%20mass%20%28LVM%29%20and%20%28b%29%20establish%20normal%20reference%20ranges%20in%20a%20selected%20group%20of%20healthy%20UK%20Biobank%20participants.%20Materials%20and%20Methods%20In%20this%20cross-sectional%20secondary%20analysis%2C%20prospectively%20collected%20data%20from%20the%20UK%20Biobank%20%282006%20to%202010%29%20were%20retrospectively%20analyzed.%20Automated%20segmentation%20of%20trabeculations%20was%20performed%20using%20a%20deep%20learning%20algorithm.%20After%20excluding%20individuals%20with%20known%20CV%20diseases%2C%20White%20adults%20without%20CV%20risk%20factors%20%28reference%20group%29%20and%20those%20with%20preexisting%20CV%20risk%20factors%20%28hypertension%2C%20hyperlipidemia%2C%20diabetes%20mellitus%2C%20or%20smoking%29%20%28exposed%20group%29%20were%20compared.%20Multivariable%20regression%20models%2C%20adjusted%20for%20potential%20confounders%20%28age%2C%20sex%2C%20and%20height%29%2C%20were%20fitted%20to%20evaluate%20the%20associations%20between%20individual%20characteristics%20and%20CV%20risk%20factors%20and%20trabeculated%20LVM.%20Results%20Of%2043%5Cu2009038%20participants%20%28mean%20age%2C%2064%20years%20%5Cu00b1%208%20%5BSD%5D%3B%2022%5Cu2009360%20women%29%2C%2028%5Cu2009672%20individuals%20%28mean%20age%2C%2066%20years%20%5Cu00b1%207%3B%2014%5Cu2009918%20men%29%20were%20included%20in%20the%20exposed%20group%2C%20and%207384%20individuals%20%28mean%20age%2C%2060%20years%20%5Cu00b1%207%3B%204729%20women%29%20were%20included%20in%20the%20reference%20group.%20Higher%20body%20mass%20index%20%28BMI%29%20%28%5Cu03b2%20%3D%200.66%20%5B95%25%20CI%3A%200.63%2C%200.68%5D%3B%20P%20%26lt%3B%20.001%29%2C%20hypertension%20%28%5Cu03b2%20%3D%200.42%20%5B95%25%20CI%3A%200.36%2C%200.48%5D%3B%20P%20%26lt%3B%20.001%29%2C%20and%20higher%20physical%20activity%20level%20%28%5Cu03b2%20%3D%200.15%20%5B95%25%20CI%3A%200.12%2C%200.17%5D%3B%20P%20%26lt%3B%20.001%29%20were%20associated%20with%20higher%20trabeculated%20LVM.%20In%20the%20reference%20group%2C%20the%20median%20trabeculated%20LVM%20was%206.3%20g%20%28IQR%2C%204.7-8.5%20g%29%20for%20men%20and%204.6%20g%20%28IQR%2C%203.4-6.0%20g%29%20for%20women.%20Median%20trabeculated%20LVM%20decreased%20with%20age%20for%20men%20from%206.5%20g%20%28IQR%2C%204.8-8.7%20g%29%20at%20age%2045-50%20years%20to%205.9%20g%20%28IQR%2C%204.3-7.8%20g%29%20at%20age%2071-80%20years%20%28P%20%3D%20.03%29.%20Conclusion%20Higher%20trabeculated%20LVM%20was%20observed%20with%20hypertension%2C%20higher%20BMI%2C%20and%20higher%20physical%20activity%20level.%20Age-%20and%20sex-specific%20reference%20ranges%20of%20trabeculated%20LVM%20in%20a%20healthy%20middle-aged%20White%20population%20were%20established.%20%5Cu00a9%20RSNA%2C%202024%20Supplemental%20material%20is%20available%20for%20this%20article.%20See%20also%20the%20editorial%20by%20Kawel-Boehm%20in%20this%20issue.%22%2C%22date%22%3A%222024-04%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1148%5C%2Fradiol.232455%22%2C%22ISSN%22%3A%221527-1315%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22KFZ6H27M%22%2C%22CXAMISN8%22%5D%2C%22dateModified%22%3A%222025-11-27T10%3A53%3A30Z%22%7D%7D%2C%7B%22key%22%3A%22DTTG9UG7%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Bartoli%20et%20al.%22%2C%22parsedDate%22%3A%222022-01%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%3BBartoli%2C%20A.%2C%20Fournel%2C%20J.%2C%20Ait-Yahia%2C%20L.%2C%20Cadour%2C%20F.%2C%20Tradi%2C%20F.%2C%20Ghattas%2C%20B.%2C%20Cortaredona%2C%20S.%2C%20Million%2C%20M.%2C%20Lasbleiz%2C%20A.%2C%20Dutour%2C%20A.%2C%20Gaborit%2C%20B.%20and%20Jacquier%2C%20A.%20%282022%29%20%26%23x201C%3BAutomatic%20Deep-Learning%20Segmentation%20of%20Epicardial%20Adipose%20Tissue%20from%20Low-Dose%20Chest%20CT%20and%20Prognosis%20Impact%20on%20COVID-19%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BCells%26lt%3B%5C%2Fi%26gt%3B%2C%2011%286%29%2C%20p.%201034.%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%2Fcells11061034%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.3390%5C%2Fcells11061034%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%22Automatic%20Deep-Learning%20Segmentation%20of%20Epicardial%20Adipose%20Tissue%20from%20Low-Dose%20Chest%20CT%20and%20Prognosis%20Impact%20on%20COVID-19%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Axel%22%2C%22lastName%22%3A%22Bartoli%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Joris%22%2C%22lastName%22%3A%22Fournel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22L%5Cu00e9a%22%2C%22lastName%22%3A%22Ait-Yahia%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Farah%22%2C%22lastName%22%3A%22Cadour%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Farouk%22%2C%22lastName%22%3A%22Tradi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Badih%22%2C%22lastName%22%3A%22Ghattas%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22S%5Cu00e9bastien%22%2C%22lastName%22%3A%22Cortaredona%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Matthieu%22%2C%22lastName%22%3A%22Million%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ad%5Cu00e8le%22%2C%22lastName%22%3A%22Lasbleiz%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Anne%22%2C%22lastName%22%3A%22Dutour%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22B%5Cu00e9n%5Cu00e9dicte%22%2C%22lastName%22%3A%22Gaborit%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Jacquier%22%7D%5D%2C%22abstractNote%22%3A%22Background%3A%20To%20develop%20a%20deep-learning%20%28DL%29%20pipeline%20that%20allowed%20an%20automated%20segmentation%20of%20epicardial%20adipose%20tissue%20%28EAT%29%20from%20low-dose%20computed%20tomography%20%28LDCT%29%20and%20investigate%20the%20link%20between%20EAT%20and%20COVID-19%20clinical%20outcomes.%20Methods%3A%20This%20monocentric%20retrospective%20study%20included%20353%20patients%3A%2095%20for%20training%2C%2020%20for%20testing%2C%20and%20238%20for%20prognosis%20evaluation.%20EAT%20segmentation%20was%20obtained%20after%20thresholding%20on%20a%20manually%20segmented%20pericardial%20volume.%20The%20model%20was%20evaluated%20with%20Dice%20coefficient%20%28DSC%29%2C%20inter-and%20intraobserver%20reproducibility%2C%20and%20clinical%20measures.%20Uni-and%20multi-variate%20analyzes%20were%20conducted%20to%20assess%20the%20prognosis%20value%20of%20the%20EAT%20volume%2C%20EAT%20extent%2C%20and%20lung%20lesion%20extent%20on%20clinical%20outcomes%2C%20including%20hospitalization%2C%20oxygen%20therapy%2C%20intensive%20care%20unit%20admission%20and%20death.%20Results%3A%20The%20mean%20DSC%20for%20EAT%20volumes%20was%200.85%20%5Cu00b1%200.05.%20For%20EAT%20volume%2C%20the%20mean%20absolute%20error%20was%2011.7%20%5Cu00b1%208.1%20cm3%20with%20a%20non-significant%20bias%20of%20%5Cu22124.0%20%5Cu00b1%2013.9%20cm3%20and%20a%20correlation%20of%200.963%20with%20the%20manual%20measures%20%28p%20%26lt%3B%200.01%29.%20The%20multivariate%20model%20providing%20the%20higher%20AUC%20to%20predict%20adverse%20outcome%20include%20both%20EAT%20extent%20and%20lung%20lesion%20extent%20%28AUC%20%3D%200.805%29.%20Conclusions%3A%20A%20DL%20algorithm%20was%20developed%20and%20evaluated%20to%20obtain%20reproducible%20and%20precise%20EAT%20segmentation%20on%20LDCT.%20EAT%20extent%20in%20association%20with%20lung%20lesion%20extent%20was%20associated%20with%20adverse%20clinical%20outcomes%20with%20an%20AUC%20%3D%200.805.%22%2C%22date%22%3A%222022%5C%2F1%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%2210.3390%5C%2Fcells11061034%22%2C%22ISSN%22%3A%222073-4409%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.mdpi.com%5C%2F2073-4409%5C%2F11%5C%2F6%5C%2F1034%22%2C%22collections%22%3A%5B%22KFZ6H27M%22%2C%22CXAMISN8%22%5D%2C%22dateModified%22%3A%222025-11-27T10%3A53%3A32Z%22%7D%7D%2C%7B%22key%22%3A%225UCW932Q%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Fournel%20et%20al.%22%2C%22parsedDate%22%3A%222021-08-12%22%2C%22numChildren%22%3A2%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%3BFournel%2C%20J.%2C%20Bartoli%2C%20A.%2C%20Bendahan%2C%20D.%2C%20Guye%2C%20M.%2C%20Bernard%2C%20M.%2C%20Rauseo%2C%20E.%2C%20Khanji%2C%20M.Y.%2C%20Petersen%2C%20S.E.%2C%20Jacquier%2C%20A.%20and%20Ghattas%2C%20B.%20%282021%29%20%26%23x201C%3BMedical%20image%20segmentation%20automatic%20quality%20control%3A%20A%20multi-dimensional%20approach%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BMedical%20Image%20Analysis%26lt%3B%5C%2Fi%26gt%3B%2C%2074%2C%20p.%20102213.%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%2F10.1016%5C%2Fj.media.2021.102213%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1016%5C%2Fj.media.2021.102213%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%22Medical%20image%20segmentation%20automatic%20quality%20control%3A%20A%20multi-dimensional%20approach%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Joris%22%2C%22lastName%22%3A%22Fournel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Axel%22%2C%22lastName%22%3A%22Bartoli%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%22Maxime%22%2C%22lastName%22%3A%22Guye%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Monique%22%2C%22lastName%22%3A%22Bernard%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Elisa%22%2C%22lastName%22%3A%22Rauseo%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Mohammed%20Y.%22%2C%22lastName%22%3A%22Khanji%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Steffen%20E.%22%2C%22lastName%22%3A%22Petersen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Jacquier%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Badih%22%2C%22lastName%22%3A%22Ghattas%22%7D%5D%2C%22abstractNote%22%3A%22In%20clinical%20applications%2C%20using%20erroneous%20segmentations%20of%20medical%20images%20can%20have%20dramatic%20consequences.%20Current%20approaches%20dedicated%20to%20medical%20image%20segmentation%20automatic%20quality%20control%20do%20not%20predict%20segmentation%20quality%20at%20slice-level%20%282D%29%2C%20resulting%20in%20sub-optimal%20evaluations.%20Our%202D-based%20deep%20learning%20method%20simultaneously%20performs%20quality%20control%20at%202D-level%20and%203D-level%20for%20cardiovascular%20MR%20image%20segmentations.%20We%20compared%20it%20with%203D%20approaches%20by%20training%20both%20on%2036%2C540%20%282D%29%20%5C%2F%203842%20%283D%29%20samples%20to%20predict%20Dice%20Similarity%20Coefficients%20%28DSC%29%20for%204%20different%20structures%20from%20the%20left%20ventricle%2C%20i.e.%2C%20trabeculations%20%28LVT%29%2C%20myocardium%20%28LVM%29%2C%20papillary%20muscles%20%28LVPM%29%20and%20blood%20%28LVC%29.%20The%202D-based%20method%20outperformed%20the%203D%20method.%20At%20the%202D-level%2C%20the%20mean%20absolute%20errors%20%28MAEs%29%20of%20the%20DSC%20predictions%20for%203823%20samples%2C%20were%200.02%2C%200.02%2C%200.05%20and%200.02%20for%20LVM%2C%20LVC%2C%20LVT%20and%20LVPM%2C%20respectively.%20At%20the%203D-level%2C%20for%20402%20samples%2C%20the%20corresponding%20MAEs%20were%200.02%2C%200.01%2C%200.02%20and%200.04.%20The%20method%20was%20validated%20in%20a%20clinical%20practice%20evaluation%20against%20semi-qualitative%20scores%20provided%20by%20expert%20cardiologists%20for%201016%20subjects%20of%20the%20UK%20BioBank.%20Finally%2C%20we%20provided%20evidence%20that%20a%20multi-level%20QC%20could%20be%20used%20to%20enhance%20clinical%20measurements%20derived%20from%20image%20segmentations.%22%2C%22date%22%3A%222021-08-12%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1016%5C%2Fj.media.2021.102213%22%2C%22ISSN%22%3A%221361-8423%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22NW6SZQBR%22%2C%22KFZ6H27M%22%2C%22XTA6KS7L%22%2C%22CXAMISN8%22%5D%2C%22dateModified%22%3A%222025-11-27T10%3A53%3A34Z%22%7D%7D%2C%7B%22key%22%3A%22SFIITJXT%22%2C%22library%22%3A%7B%22id%22%3A180164%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Bartoli%20et%20al.%22%2C%22parsedDate%22%3A%222021-01%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%3BBartoli%2C%20A.%2C%20Fournel%2C%20J.%2C%20Bentatou%2C%20Z.%2C%20Habib%2C%20G.%2C%20Lalande%2C%20A.%2C%20Bernard%2C%20M.%2C%20Boussel%2C%20L.%2C%20Pontana%2C%20F.%2C%20Dacher%2C%20J.-N.%2C%20Ghattas%2C%20B.%20and%20Jacquier%2C%20A.%20%282021%29%20%26%23x201C%3BDeep%20Learning-based%20Automated%20Segmentation%20of%20Left%20Ventricular%20Trabeculations%20and%20Myocardium%20on%20Cardiac%20MR%20Images%3A%20A%20Feasibility%20Study%2C%26%23x201D%3B%20%26lt%3Bi%26gt%3BRadiology.%20Artificial%20Intelligence%26lt%3B%5C%2Fi%26gt%3B%2C%203%281%29%2C%20p.%20e200021.%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%2F10.1148%5C%2Fryai.2020200021%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1148%5C%2Fryai.2020200021%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%22Deep%20Learning-based%20Automated%20Segmentation%20of%20Left%20Ventricular%20Trabeculations%20and%20Myocardium%20on%20Cardiac%20MR%20Images%3A%20A%20Feasibility%20Study%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Axel%22%2C%22lastName%22%3A%22Bartoli%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Joris%22%2C%22lastName%22%3A%22Fournel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zakarya%22%2C%22lastName%22%3A%22Bentatou%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Gilbert%22%2C%22lastName%22%3A%22Habib%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alain%22%2C%22lastName%22%3A%22Lalande%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Monique%22%2C%22lastName%22%3A%22Bernard%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Lo%5Cu00efc%22%2C%22lastName%22%3A%22Boussel%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fran%5Cu00e7ois%22%2C%22lastName%22%3A%22Pontana%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jean-Nicolas%22%2C%22lastName%22%3A%22Dacher%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Badih%22%2C%22lastName%22%3A%22Ghattas%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alexis%22%2C%22lastName%22%3A%22Jacquier%22%7D%5D%2C%22abstractNote%22%3A%22Purpose%3A%20To%20develop%20and%20evaluate%20a%20complete%20deep%20learning%20pipeline%20that%20allows%20fully%20automated%20end-diastolic%20left%20ventricle%20%28LV%29%20cardiac%20MRI%20segmentation%2C%20including%20trabeculations%20and%20automatic%20quality%20control%20of%20the%20predicted%20segmentation.%5CnMaterials%20and%20Methods%3A%20This%20multicenter%20retrospective%20study%20includes%20training%2C%20validation%2C%20and%20testing%20datasets%20of%20272%2C%2027%2C%20and%20150%20cardiac%20MR%20images%2C%20respectively%2C%20collected%20between%202012%20and%202018.%20The%20reference%20standard%20was%20the%20manual%20segmentation%20of%20four%20LV%20anatomic%20structures%20performed%20on%20end-diastolic%20short-axis%20cine%20cardiac%20MRI%3A%20LV%20trabeculations%2C%20LV%20myocardium%2C%20LV%20papillary%20muscles%2C%20and%20the%20LV%20blood%20cavity.%20The%20automatic%20pipeline%20was%20composed%20of%20five%20steps%20with%20a%20DenseNet%20architecture.%20Intraobserver%20agreement%2C%20interobserver%20agreement%2C%20and%20interaction%20time%20were%20recorded.%20The%20analysis%20includes%20the%20correlation%20between%20the%20manual%20and%20automated%20segmentation%2C%20a%20reproducibility%20comparison%2C%20and%20Bland-Altman%20plots.%5CnResults%3A%20The%20automated%20method%20achieved%20mean%20Dice%20coefficients%20of%200.96%20%5Cu00b1%200.01%20%28standard%20deviation%29%20for%20LV%20blood%20cavity%2C%200.89%20%5Cu00b1%200.03%20for%20LV%20myocardium%2C%20and%200.62%20%5Cu00b1%200.08%20for%20LV%20trabeculation%20%28mean%20absolute%20error%2C%203.63%20g%20%5Cu00b1%203.4%29.%20Automatic%20quantification%20of%20LV%20end-diastolic%20volume%2C%20LV%20myocardium%20mass%2C%20LV%20trabeculation%2C%20and%20trabeculation%20mass-to-total%20myocardial%20mass%20%28TMM%29%20ratio%20showed%20a%20significant%20correlation%20with%20the%20manual%20measures%20%28r%20%3D%200.99%2C%200.99%2C%200.90%2C%20and%200.83%2C%20respectively%3B%20all%20P%20%26lt%3B%20.01%29.%20On%20a%20subset%20of%2048%20patients%2C%20the%20mean%20Dice%20value%20for%20LV%20trabeculation%20was%200.63%20%5Cu00b1%200.10%20or%20higher%20compared%20with%20the%20human%20interobserver%20%280.44%20%5Cu00b1%200.09%3B%20P%20%26lt%3B%20.01%29%20and%20intraobserver%20measures%20%280.58%20%5Cu00b1%200.09%3B%20P%20%26lt%3B%20.01%29.%20Automatic%20quantification%20of%20the%20trabeculation%20mass-to-TMM%20ratio%20had%20a%20higher%20correlation%20%280.92%29%20compared%20with%20the%20intra-%20and%20interobserver%20measures%20%280.74%20and%200.39%2C%20respectively%3B%20both%20P%20%26lt%3B%20.01%29.%5CnConclusion%3A%20Automated%20deep%20learning%20framework%20can%20achieve%20reproducible%20and%20quality-controlled%20segmentation%20of%20cardiac%20trabeculations%2C%20outperforming%20inter-%20and%20intraobserver%20analyses.Supplemental%20material%20is%20available%20for%20this%20article.%5Cu00a9%5Cu2009RSNA%2C%202020.%22%2C%22date%22%3A%222021-01%22%2C%22language%22%3A%22eng%22%2C%22DOI%22%3A%2210.1148%5C%2Fryai.2020200021%22%2C%22ISSN%22%3A%222638-6100%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%225Z6CVJGB%22%2C%22KFZ6H27M%22%2C%22B4FUISBZ%22%2C%22CXAMISN8%22%5D%2C%22dateModified%22%3A%222025-11-27T10%3A53%3A33Z%22%7D%7D%5D%7D
Fournel, J., De Masi, M., Lu, C., Omnes, V., Muselier, B., Ghattas, B., Bouchot, O., Kafi, M., Lalande, A., Gaudry, M., Jacquier, A. and Bartoli, A. (2025) “Deep learning-driven false-lumen volumes predict adverse remodeling better than diameter in patients with residual aortic dissection on CT,” European Radiology [Preprint]. Available at: https://doi.org/10.1007/s00330-025-12116-9.
Gaborit, B., Julla, J.B., Fournel, J., Ancel, P., Soghomonian, A., Deprade, C., Lasbleiz, A., Houssays, M., Ghattas, B., Gascon, P., Righini, M., Matonti, F., Venteclef, N., Potier, L., Gautier, J.F., Resseguier, N., Bartoli, A., Mourre, F., Darmon, P., Jacquier, A. and Dutour, A. (2024) “Fully automated epicardial adipose tissue volume quantification with deep learning and relationship with CAC score and micro/macrovascular complications in people living with type 2 diabetes: the multicenter EPIDIAB study,” Cardiovascular Diabetology, 23(1), p. 328. Available at: https://doi.org/10.1186/s12933-024-02411-y.
Aung, N., Bartoli, A., Rauseo, E., Cortaredona, S., Sanghvi, M.M., Fournel, J., Ghattas, B., Khanji, M.Y., Petersen, S.E. and Jacquier, A. (2024) “Left Ventricular Trabeculations at Cardiac MRI: Reference Ranges and Association with Cardiovascular Risk Factors in UK Biobank,” Radiology, 311(1), p. e232455. Available at: https://doi.org/10.1148/radiol.232455.
Bartoli, A., Fournel, J., Ait-Yahia, L., Cadour, F., Tradi, F., Ghattas, B., Cortaredona, S., Million, M., Lasbleiz, A., Dutour, A., Gaborit, B. and Jacquier, A. (2022) “Automatic Deep-Learning Segmentation of Epicardial Adipose Tissue from Low-Dose Chest CT and Prognosis Impact on COVID-19,” Cells, 11(6), p. 1034. Available at: https://doi.org/10.3390/cells11061034.
Fournel, J., Bartoli, A., Bendahan, D., Guye, M., Bernard, M., Rauseo, E., Khanji, M.Y., Petersen, S.E., Jacquier, A. and Ghattas, B. (2021) “Medical image segmentation automatic quality control: A multi-dimensional approach,” Medical Image Analysis, 74, p. 102213. Available at: https://doi.org/10.1016/j.media.2021.102213.
Bartoli, A., Fournel, J., Bentatou, Z., Habib, G., Lalande, A., Bernard, M., Boussel, L., Pontana, F., Dacher, J.-N., Ghattas, B. and Jacquier, A. (2021) “Deep Learning-based Automated Segmentation of Left Ventricular Trabeculations and Myocardium on Cardiac MR Images: A Feasibility Study,” Radiology. Artificial Intelligence, 3(1), p. e200021. Available at: https://doi.org/10.1148/ryai.2020200021.