Deep Learning for Multi-Tissue Segmentation and Fully Automatic Personalized Biomechanical Models from BACPAC Clinical Lumbar Spine MRI

被引:11
|
作者
Hess, Madeline [1 ,3 ]
Allaire, Brett [2 ]
Gao, Kenneth T. [1 ]
Tibrewala, Radhika [1 ]
Inamdar, Gaurav [1 ]
Bharadwaj, Upasana [1 ]
Chin, Cynthia [1 ]
Pedoia, Valentina [1 ]
Bouxsein, Mary [2 ]
Anderson, Dennis [2 ]
Majumdar, Sharmila [1 ]
机构
[1] Univ Calif San Francisco, Ctr Intelligent Imaging, Dept Radiol & Biomed Imaging, San Francisco, CA USA
[2] Harvard Med Sch, Ctr Adv Orthoped Studies, Beth Israel Deaconess Med Ctr, Boston, MA USA
[3] UCSF, Ctr Intelligent Imaging, Dept Radiol & Biomed Imaging, Box 2050,1700 4th St, San Francisco, CA 94158 USA
基金
美国国家卫生研究院;
关键词
Deep Learning; Magnetic Resonance Imaging; Musculoskeletal; Biomechanics; Quantitative Imaging; Lumbar Spine; Chronic Low Back Pain; BACPAC; LOW-BACK-PAIN; CROSS-SECTIONAL AREA; INTERVERTEBRAL DISC HEIGHT; PREVALENCE; COSTS;
D O I
10.1093/pm/pnac142
中图分类号
R614 [麻醉学];
学科分类号
100217 ;
摘要
Study Design In vivo retrospective study of fully automatic quantitative imaging feature extraction from clinically acquired lumbar spine magnetic resonance imaging (MRI). Objective To demonstrate the feasibility of substituting automatic for human-demarcated segmentation of major anatomic structures in clinical lumbar spine MRI to generate quantitative image-based features and biomechanical models. Setting Previous studies have demonstrated the viability of automatic segmentation applied to medical images; however, the feasibility of these networks to segment clinically acquired images has not yet been demonstrated, as they largely rely on specialized sequences or strict quality of imaging data to achieve good performance. Methods Convolutional neural networks were trained to demarcate vertebral bodies, intervertebral disc, and paraspinous muscles from sagittal and axial T1-weighted MRIs. Intervertebral disc height, muscle cross-sectional area, and subject-specific musculoskeletal models of tissue loading in the lumbar spine were then computed from these segmentations and compared against those computed from human-demarcated masks. Results Segmentation masks, as well as the morphological metrics and biomechanical models computed from those masks, were highly similar between human- and computer-generated methods. Segmentations were similar, with Dice similarity coefficients of 0.77 or greater across networks, and morphological metrics and biomechanical models were similar, with Pearson R correlation coefficients of 0.69 or greater when significant. Conclusions This study demonstrates the feasibility of substituting computer-generated for human-generated segmentations of major anatomic structures in lumbar spine MRI to compute quantitative image-based morphological metrics and subject-specific musculoskeletal models of tissue loading quickly, efficiently, and at scale without interrupting routine clinical care.
引用
收藏
页码:S139 / S148
页数:10
相关论文
共 50 条
  • [31] Development of lumbar spine MRI referrals vetting models using machine learning and deep learning algorithms: Comparison models vs healthcare professionals
    Alanazi, A. H.
    Cradock, A.
    Rainford, L.
    RADIOGRAPHY, 2022, 28 (03) : 674 - 683
  • [32] Automatic semantic segmentation of the lumbar spine: Clinical applicability in a multi-parametric and multi-center study on magnetic resonance images
    Saenz-Gamboa, Jhon Jairo
    Domenech, Julio
    Alonso-Manjarres, Antonio
    Gomez, Jon A.
    Iglesia-Vaya, Maria de la
    ARTIFICIAL INTELLIGENCE IN MEDICINE, 2023, 140
  • [33] Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?
    Bernard, Olivier
    Lalande, Alain
    Zotti, Clement
    Cervenansky, Frederick
    Yang, Xin
    Heng, Pheng-Ann
    Cetin, Irem
    Lekadir, Karim
    Camara, Oscar
    Gonzalez Ballester, Miguel Angel
    Sanroma, Gerard
    Napel, Sandy
    Petersen, Steffen
    Tziritas, Georgios
    Grinias, Elias
    Khened, Mahendra
    Kollerathu, Varghese Alex
    Krishnamurthi, Ganapathy
    Rohe, Marc-Michel
    Pennec, Xavier
    Sermesant, Maxime
    Isensee, Fabian
    Jaeger, Paul
    Maier-Hein, Klaus H.
    Full, Peter M.
    Wolf, Ivo
    Engelhardt, Sandy
    Baumgartner, Christian F.
    Koch, Lisa M.
    Wolterink, Jelmer M.
    Isgum, Ivana
    Jang, Yeonggul
    Hong, Yoonmi
    Patravali, Jay
    Jain, Shubham
    Humbert, Olivier
    Jodoin, Pierre-Marc
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2018, 37 (11) : 2514 - 2525
  • [34] A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI
    Avendi, M. R.
    Kheradvar, Arash
    Jafarkhani, Hamid
    MEDICAL IMAGE ANALYSIS, 2016, 30 : 108 - 119
  • [35] Priority attention network with Bayesian learning for fully automatic segmentation of substantia nigra from neuromelanin MRI
    Hu, Tao
    Itoh, Hayato
    Oda, Masahiro
    Saiki, Shinji
    Hattori, Nobutaka
    Kamagata, Koji
    Aoki, Shigeki
    Mori, Kensaku
    MEDICAL IMAGING 2023, 2023, 12464
  • [36] Estimating lumbar bone mineral density from conventional MRI and radiographs with deep learning in spine patients
    Galbusera, Fabio
    Cina, Andrea
    O'Riordan, Dave
    Vitale, Jacopo A.
    Loibl, Markus
    Fekete, Tamas F.
    Kleinstueck, Frank
    Haschtmann, Daniel
    Mannion, Anne F.
    EUROPEAN SPINE JOURNAL, 2024, 33 (11) : 4092 - 4103
  • [37] Automatic Spine Tissue Segmentation from MRI Data Based on Cascade of Boosted Classifiers and Active Appearance Model
    Gawel, Dominik
    Glowka, Paweb
    Kotwicki, Tomasz
    Nowak, Michal
    BIOMED RESEARCH INTERNATIONAL, 2018, 2018
  • [38] Deep learning-based segmentation of knee MRI for fully automatic subregional morphological assessment of cartilage tissues: Data from the Osteoarthritis Initiative
    Panfilov, Egor
    Tiulpin, Aleksei
    Nieminen, Miika T.
    Saarakkala, Simo
    Casula, Victor
    JOURNAL OF ORTHOPAEDIC RESEARCH, 2022, 40 (05) : 1113 - 1124
  • [39] Automatic deep learning-based myocardial infarction segmentation from delayed enhancement MRI
    Chen, Zhihao
    Lalande, Alain
    Salomon, Michel
    Decourselle, Thomas
    Pommier, Thibaut
    Qayyum, Abdul
    Shi, Jixi
    Perrot, Gilles
    Couturier, Raphaël
    Computerized Medical Imaging and Graphics, 2022, 95
  • [40] Automatic deep learning-based myocardial infarction segmentation from delayed enhancement MRI
    Chen, Zhihao
    Lalande, Alain
    Salomon, Michel
    Decourselle, Thomas
    Pommier, Thibaut
    Qayyum, Abdul
    Shi, Jixi
    Perrot, Gilles
    Couturier, Raphael
    COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2022, 95