دورية أكاديمية
أعرض في EDS

Deep learning-based automatic pipeline for quantitative assessment of thigh muscle morphology and fatty infiltration.

التفاصيل البيبلوغرافية
العنوان: Deep learning-based automatic pipeline for quantitative assessment of thigh muscle morphology and fatty infiltration.
المؤلفون: Gaj S; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Eck BL; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA.; Department of Radiology, Imaging Institute, Cleveland Clinic, Cleveland, Ohio, USA., Xie D; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Lartey R; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Lo C; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Zaylor W; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Yang M; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Nakamura K; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA., Winalski CS; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA.; Department of Radiology, Imaging Institute, Cleveland Clinic, Cleveland, Ohio, USA., Spindler KP; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Orthopaedics, Cleveland Clinic Florida Region, Weston, Florida, USA., Li X; Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, Ohio, USA.; Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA.; Department of Radiology, Imaging Institute, Cleveland Clinic, Cleveland, Ohio, USA.
المصدر: Magnetic resonance in medicine [Magn Reson Med] 2023 Jun; Vol. 89 (6), pp. 2441-2455. Date of Electronic Publication: 2023 Feb 06.
نوع المنشور: Journal Article; Research Support, N.I.H., Extramural
اللغة: English
بيانات الدورية: Publisher: Wiley Country of Publication: United States NLM ID: 8505245 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1522-2594 (Electronic) Linking ISSN: 07403194 NLM ISO Abbreviation: Magn Reson Med Subsets: MEDLINE
أسماء مطبوعة: Publication: 1999- : New York, NY : Wiley
Original Publication: San Diego : Academic Press,
مواضيع طبية MeSH: Thigh*/diagnostic imaging , Deep Learning*, Humans ; Reproducibility of Results ; Knee Joint ; Muscle, Skeletal/diagnostic imaging ; Magnetic Resonance Imaging/methods
مستخلص: Purpose: Fast and accurate thigh muscle segmentation from MRI is important for quantitative assessment of thigh muscle morphology and composition. A novel deep learning (DL) based thigh muscle and surrounding tissues segmentation model was developed for fully automatic and reproducible cross-sectional area (CSA) and fat fraction (FF) quantification and tested in patients at 10 years after anterior cruciate ligament reconstructions.
Methods: A DL model combining UNet and DenseNet was trained and tested using manually segmented thighs from 16 patients (32 legs). Segmentation accuracy was evaluated using Dice similarity coefficients (DSC) and average symmetric surface distance (ASSD). A UNet model was trained for comparison. These segmentations were used to obtain CSA and FF quantification. Reproducibility of CSA and FF quantification was tested with scan and rescan of six healthy subjects.
Results: The proposed UNet and DenseNet had high agreement with manual segmentation (DSC >0.97, ASSD < 0.24) and improved performance compared with UNet. For hamstrings of the operated knee, the automated pipeline had largest absolute difference of 6.01% for CSA and 0.47% for FF as compared to manual segmentation. In reproducibility analysis, the average difference (absolute) in CSA quantification between scan and rescan was better for the automatic method as compared with manual segmentation (2.27% vs. 3.34%), whereas the average difference (absolute) in FF quantification were similar.
Conclusions: The proposed method exhibits excellent accuracy and reproducibility in CSA and FF quantification compared with manual segmentation and can be used in large-scale patient studies.
(© 2023 The Authors. Magnetic Resonance in Medicine published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine.)
References: Am J Sports Med. 2014 Feb;42(2):302-11. (PMID: 24275860)
Ann Rheum Dis. 2004 Mar;63(3):269-73. (PMID: 14962961)
Lancet Neurol. 2016 Jan;15(1):65-77. (PMID: 26549782)
Radiol Artif Intell. 2020 Oct 28;2(6):e200010. (PMID: 33937847)
J Cachexia Sarcopenia Muscle. 2018 Dec;9(6):1093-1100. (PMID: 30221479)
Magn Reson Med. 2018 Dec;80(6):2759-2770. (PMID: 29774599)
MAGMA. 2017 Oct;30(5):489-503. (PMID: 28455629)
MAGMA. 2016 Apr;29(2):245-57. (PMID: 26983429)
Eur J Radiol. 2018 Jun;103:57-64. (PMID: 29803386)
J Orthop Res. 2016 Nov;34(11):1843-1855. (PMID: 27612195)
J Orthop Trauma. 2006 Nov-Dec;20(10):739-44. (PMID: 17106388)
Br J Radiol. 2018 Sep;91(1089):20180252. (PMID: 30004791)
Magn Reson Med. 2018 Apr;79(4):2379-2391. (PMID: 28733975)
Magn Reson Med. 2020 Jul;84(1):437-449. (PMID: 31793071)
Insights Imaging. 2020 Nov 30;11(1):128. (PMID: 33252711)
Osteoarthritis Cartilage. 2022 Dec;30(12):1647-1657. (PMID: 36049665)
Eur Radiol Exp. 2019 Nov 20;3(1):45. (PMID: 31748839)
Med Image Comput Comput Assist Interv. 2012;15(Pt 1):569-76. (PMID: 23285597)
PLoS One. 2015 Jul 16;10(7):e0132717. (PMID: 26181385)
Arch Orthop Trauma Surg. 2014 Oct;134(10):1417-28. (PMID: 25091127)
Magn Reson Med. 2019 Apr;81(4):2736-2745. (PMID: 30311704)
Acta Biomater. 2022 Nov;153:243-259. (PMID: 36374749)
J Bone Joint Surg Am. 2012 Jul 18;94(14):1274-82. (PMID: 22810397)
Radiology. 2018 Jul;288(1):177-185. (PMID: 29584598)
Magn Reson Med. 2020 Apr;83(4):1471-1483. (PMID: 31631409)
J Magn Reson Imaging. 2015 Jun;41(6):1558-69. (PMID: 25111561)
J Magn Reson Imaging. 2016 Mar;43(3):601-10. (PMID: 26268693)
Osteoarthritis Cartilage. 2014 Feb;22(2):226-34. (PMID: 24361743)
Neuroimage. 2011 Feb 1;54(3):2033-44. (PMID: 20851191)
J Magn Reson Imaging. 2021 May;53(5):1539-1549. (PMID: 33448058)
Osteoarthritis Cartilage. 2018 Sep;26(9):1190-1195. (PMID: 29890261)
Exerc Sport Sci Rev. 2009 Jul;37(3):147-53. (PMID: 19550206)
PLoS One. 2018 Jun 7;13(6):e0198200. (PMID: 29879128)
MAGMA. 2020 Aug;33(4):483-493. (PMID: 31872357)
PLoS One. 2014 Feb 28;9(2):e90377. (PMID: 24587344)
معلومات مُعتمدة: K25 AR078928 United States AR NIAMS NIH HHS; R01 AR075422 United States AR NIAMS NIH HHS; T32 AR007505 United States AR NIAMS NIH HHS
فهرسة مساهمة: Keywords: automated segmentation; deep learning; magnetic resonance imaging; thigh muscle
تواريخ الأحداث: Date Created: 20230206 Date Completed: 20230328 Latest Revision: 20240603
رمز التحديث: 20240603
مُعرف محوري في PubMed: PMC10050107
DOI: 10.1002/mrm.29599
PMID: 36744695
قاعدة البيانات: MEDLINE
الوصف
تدمد:1522-2594
DOI:10.1002/mrm.29599