K Linka1, J Thüring2, L Rieppo3, R C Aydin4, C J Cyron5, C Kuhl6, D Merhof7, D Truhn8, S Nebelung9. 1. Institute of Continuum and Materials Mechanics, Hamburg University of Technology, Hamburg, 21073, Germany. Electronic address: kevin.linka@tuhh.de. 2. Department of Diagnostic and Interventional Radiology, Aachen University Hospital, Aachen, 52074, Germany. Electronic address: jthuering@ukaachen.de. 3. Research Unit of Medical Imaging, Physics and Technology, Faculty of Medicine, University of Oulu, Finland. Electronic address: lassi.rieppo@oulu.fi. 4. Institute of Materials Research, Materials Mechanics, Helmholtz-Zentrum Geesthacht, Geesthacht, Germany. Electronic address: Roland.Aydin@hzg.de. 5. Institute of Continuum and Materials Mechanics, Hamburg University of Technology, Hamburg, 21073, Germany; Institute of Materials Research, Materials Mechanics, Helmholtz-Zentrum Geesthacht, Geesthacht, Germany. Electronic address: christian.cyron@tuhh.de. 6. Department of Diagnostic and Interventional Radiology, Aachen University Hospital, Aachen, 52074, Germany. Electronic address: ckuhl@ukaachen.de. 7. Institute of Imaging and Computer Vision, RWTH Aachen University, Aachen, 52074, Germany. Electronic address: dorit.merhof@lfb.rwth-aachen.de. 8. Department of Diagnostic and Interventional Radiology, Aachen University Hospital, Aachen, 52074, Germany; Institute of Imaging and Computer Vision, RWTH Aachen University, Aachen, 52074, Germany. Electronic address: dtruhn@ukaachen.de. 9. Department of Diagnostic and Interventional Radiology, Medical Faculty, University of Dusseldorf, 40225, Dusseldorf, Germany. Electronic address: snebelung@ukaachen.de.
Abstract
BACKGROUND: Articular cartilage degeneration is the hallmark change of osteoarthritis, a severely disabling disease with high prevalence and considerable socioeconomic and individual burden. Early, potentially reversible cartilage degeneration is characterized by distinct changes in cartilage composition and ultrastructure, while the tissue's morphology remains largely unaltered. Hence, early degenerative changes may not be diagnosed by clinical standard diagnostic tools. METHODS: Against this background, this study introduces a novel method to determine the tissue composition non-invasively. Our method involves quantitative MRI parameters (i.e., T1, T1ρ, T2 and [Formula: see text] maps), compositional reference measurements (i.e., microspectroscopically determined local proteoglycan [PG] and collagen [CO] contents) and machine learning techniques (i.e., artificial neural networks [ANNs] and multivariate linear models [MLMs]) on 17 histologically grossly intact human cartilage samples. RESULTS: Accuracy and precision were higher in ANN-based predictions than in MLM-based predictions and moderate-to-strong correlations were found between measured and predicted compositional parameters. CONCLUSION: Once trained for the clinical setting, advanced machine learning techniques, in particular ANNs, may be used to non-invasively determine compositional features of cartilage based on quantitative MRI parameters with potential implications for the diagnosis of (early) degeneration and for the monitoring of therapeutic outcomes.
BACKGROUND: Articular cartilage degeneration is the hallmark change of osteoarthritis, a severely disabling disease with high prevalence and considerable socioeconomic and individual burden. Early, potentially reversible cartilage degeneration is characterized by distinct changes in cartilage composition and ultrastructure, while the tissue's morphology remains largely unaltered. Hence, early degenerative changes may not be diagnosed by clinical standard diagnostic tools. METHODS: Against this background, this study introduces a novel method to determine the tissue composition non-invasively. Our method involves quantitative MRI parameters (i.e., T1, T1ρ, T2 and [Formula: see text] maps), compositional reference measurements (i.e., microspectroscopically determined local proteoglycan [PG] and collagen [CO] contents) and machine learning techniques (i.e., artificial neural networks [ANNs] and multivariate linear models [MLMs]) on 17 histologically grossly intact human cartilage samples. RESULTS: Accuracy and precision were higher in ANN-based predictions than in MLM-based predictions and moderate-to-strong correlations were found between measured and predicted compositional parameters. CONCLUSION: Once trained for the clinical setting, advanced machine learning techniques, in particular ANNs, may be used to non-invasively determine compositional features of cartilage based on quantitative MRI parameters with potential implications for the diagnosis of (early) degeneration and for the monitoring of therapeutic outcomes.
Authors: Marc Sebastian Huppertz; Justus Schock; Karl Ludger Radke; Daniel Benjamin Abrar; Manuel Post; Christiane Kuhl; Daniel Truhn; Sven Nebelung Journal: Life (Basel) Date: 2021-03-05