CoChat
Sign Up
Cited by 63Radiomics and Machine Learning in Medical ImagingOsteoarthritis Treatment and MechanismsImage Retrieval and Classification Techniques

A machine learning approach to distinguish between knees without and with osteoarthritis using MRI-based radiomic features from tibial bone

Open access
Hirvasniemi, Jukka, Klein, Stefan, Bierma-Zeinstra, Sita, Vernooij, Meike W., Schiphof, Dieuwke, Oei, Edwin H. G.·2021·European Radiology
DOI 10.1007/s00330-021-07951-5PMC PMC8523397
Open this paper in CoChat — free accountOpen PDF

Abstract

Objectives: Our aim was to assess the ability of semi-automatically extracted magnetic resonance imaging (MRI)–based radiomic features from tibial subchondral bone to distinguish between knees without and with osteoarthritis. Methods: The right knees of 665 females from the population-based Rotterdam Study scanned with 1.5T MRI were analyzed. A fast imaging employing steady-state acquisition sequence was used for the quantitative bone analyses. Tibial bone was segmented using a method that combines multi-atlas and appearance models. Radiomic features related to the shape and texture were calculated from six volumes of interests (VOIs) in the proximal tibia. Machine learning–based Elastic Net models with 10-fold cross-validation were used to distinguish between knees without and with MRI Osteoarthritis Knee Score (MOAKS)–based tibiofemoral osteoarthritis. Performance of the covariate (age and body mass index), image features, and combined covariate + image features models were assessed using the area under the receiver operating characteristic curve (ROC AUC). Results: Of 665 analyzed knees, 76 (11.4%) had osteoarthritis. An ROC AUC of 0.68 (95% confidence interval (CI): 0.60–0.75) was obtained using the covariate model. The image features model yielded an ROC AUC of 0.80 (CI: 0.73–0.87). The model that combined image features from all VOIs and covariates yielded an ROC AUC of 0.80 (CI: 0.73–0.87). Conclusion: Our results suggest that radiomic features are useful imaging biomarkers of subchondral bone for the diagnosis of osteoarthritis. An advantage of assessing bone on MRI instead of on radiographs is that other tissues can be assessed simultaneously. Key Points: • Subchondral bone plays a role in the osteoarthritis disease processes. • MRI radiomics is a potential method for quantifying changes in subchondral bone. • Semi-automatically extracted radiomic features of tibia differ between subjects without and with osteoarthritis.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1007/s00330-021-07951-5

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Together, these approaches enable systems-level insights into disease mechanisms, improve prognostic stratification, and support the development of standardized, clinically actionable biomarkers for degenerative joint disease management.

    “Together, these approaches enable systems-level insights into disease mechanisms, improve prognostic stratification, and support the development of standardized, clinically actionable biomarkers for degenerative joint disease management, despite challenges like technical complexity, data integration, and evolving assay technologies [79].”

    From Biomarker-Guided Imaging and AI-Augmented Diagnosis of Degenerative Joint Disease (2025) · cites this paper as [79]

    The Discussion (page unavailable) says tibial-bone radiomic features could serve as quantitative imaging biomarkers in future studies and could contribute to a more comprehensive view of osteoarthritis-related tissue changes. It does not establish prognostic stratification, standardized clinically actionable biomarkers, or systems-level insight into disease mechanisms; it says prognostic value still needs study and the link between features and biological processes is not always clear. The cited technical, data-integration, and assay-technology challenges are also not supported by these passages. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • Radiomics can transform imaging into high-dimensional data capturing subtle patterns in joint structures such as subchondral bone.

    “Through image acquisition, preprocessing, region of interest (ROI) segmentation, feature extraction (first-order statistical, shape-based, second-order texture via GLCM/GLRLM, higher-order filtered features), feature selection, and model development, radiomics can essentially help transform imaging into high-dimensional data, capturing subtle patterns in joint structures like subchondral bone [79,”

    From Biomarker-Guided Imaging and AI-Augmented Diagnosis of Degenerative Joint Disease (2025) · cites this paper as [79]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

The citation lists haven’t been fetched yet

References 50 · Cited by 63

Research works printed but not linked: 45

  1. [1]Conaghan PG Kloppenburg M Schett G Bijlsma JWJ Osteoarthritis research priorities: a report from a EULAR ad hoc expert committee Ann Rheum Dis 2014 73 1442 1445 24625626
  2. [2]Salmon JH Rat AC Sellam J Economic impact of lower-limb osteoarthritis worldwide: a systematic review of cost-of-illness studies Osteoarthritis Cartilage 2016 24 1500 1508 27034093
  3. [3]Hunter DJ Schofield D Callander E The individual and socioeconomic impact of osteoarthritis Nat Rev Rheumatol 2014 10 437 441 24662640
  4. [4]Buckwalter JA Mankin HJ Articular cartilage: degeneration and osteoarthritis, repair, regeneration, and transplantation Instr Course Lect 1998 47 487 504 9571450
  5. [5]Weinans H Siebelt M Agricola R Botter SM Piscaer TM Waarsing JH Pathophysiology of peri-articular bone changes in osteoarthritis Bone 2012 51 190 196 22343134
  6. [6]Radin EL, Rose RM (1986) Role of subchondral bone in the initiation and progression of cartilage damage. Clin Orthop Relat Res 34–40 3780104
  7. [7]Mansell JP Collins C Bailey AJ Bone, not cartilage, should be the major focus in osteoarthritis Nat Clin Pract Rheumatol 2007 3 306 307 17538562
  8. [8]Kwan Tat S Lajeunesse D Pelletier JP Martel-Pelletier J Targeting subchondral bone for treating osteoarthritis: what is the evidence? Best Pract Res Clin Rheumatol 2010 24 51 70 20129200 PMC5250505
  9. [9]Roemer FW Crema MD Trattnig S Guermazi A Advances in imaging of osteoarthritis and cartilage Radiology 2011 260 332 354 21778451
  10. [10]Hunter DJ Guermazi A Lo GH Evolution of semi-quantitative whole joint assessment of knee OA: MOAKS (MRI Osteoarthritis Knee Score) Osteoarthritis Cartilage 2011 19 990 1002 21645627 PMC4058435
  11. [11]Oei EHG van Tiel J Robinson WH Gold GE Quantitative radiologic imaging techniques for articular cartilage composition: toward early diagnosis and development of disease-modifying therapeutics for osteoarthritis Arthritis Care Res 2014 66 1129 1141 PMC4321941 24578345
  12. [12]Pedoia V Majumdar S Translation of morphological and functional musculoskeletal imaging J Orthop Res 2019 37 23 34 30273968
  13. [13]Roemer FW Demehri S Omoumi P State of the Art: Imaging of Osteoarthritis–Revisited 2020 Radiology 2020 296 5 21 32427556
  14. [14]Starmans MPA, van der Voort SR, Tovar JMC, Veenland JF, Klein S, Niessen WJ (2020) Radiomics: data mining using quantitative medical image features. Handbook of Medical Image Computing and Computer Assisted Intervention. Elsevier, pp 429–456
  15. [15]MacKay JW Kapoor G Driban JB Association of subchondral bone texture on magnetic resonance imaging with radiographic knee osteoarthritis progression: data from the Osteoarthritis Initiative Bone Ancillary Study Eur Radiol 2018 28 4687 4695 29721684 PMC6182744
  16. [16]MacKay JW Murray PJ Kasmai B Johnson G Donell ST Toms AP MRI texture analysis of subchondral bone at the tibial plateau Eur Radiol 2016 26 3034 3045 26679180
  17. [17]MacKay JW Murray PJ Low SBL Quantitative analysis of tibial subchondral bone: texture analysis outperforms conventional trabecular microarchitecture analysis J Magn Reson Imaging 2016 43 1159 1170 26606692
  18. [18]Bowes MA, Kacena K, Alabas OA et al (2020) Machine-learning, MRI bone shape and important clinical outcomes in osteoarthritis: data from the Osteoarthritis Initiative. Ann Rheum Dis. 10.1136/annrheumdis-2020-217160 PMC7958089 33188042 · DOI
  19. [19]MacKay JW Murray PJ Kasmai B Johnson G Donell ST Toms AP Subchondral bone in osteoarthritis: association between MRI texture analysis and histomorphometry Osteoarthritis Cartilage 2017 25 700 707 27986620
  20. [20]Schiphof D van Middelkoop M de Klerk BM Crepitus is a first indication of patellofemoral osteoarthritis (and not of tibiofemoral osteoarthritis) Osteoarthritis Cartilage 2014 22 631 638 24583066
  21. [21]Schiphof D Oei EHG Hofman A Waarsing JH Weinans H Bierma-Zeinstra SMA Sensitivity and associations with pain and body weight of an MRI definition of knee osteoarthritis compared with radiographic Kellgren and Lawrence criteria: a population-based study in middle-aged females Osteoarthritis Cartilage 2014 22 440 446 24418678
  22. [22]Ikram MA Brusselle GGO Murad SD The Rotterdam Study: 2018 update on objectives, design and main results Eur J Epidemiol 2017 32 807 850 29064009 PMC5662692
  23. [23]Runhaar J Schiphof D van Meer B Reijman M Bierma-Zeinstra SMA Oei EHG How to define subregional osteoarthritis progression using semi-quantitative MRI Osteoarthritis Knee Score (MOAKS) Osteoarthritis Cartilage 2014 22 1533 1536 25278062
  24. [24]Hunter DJ Arden N Conaghan PG Definition of osteoarthritis on MRI: results of a Delphi exercise Osteoarthritis Cartilage 2011 19 963 969 21620986 PMC3261513
  25. [25]Emery CA Whittaker JL Mahmoudian A Establishing outcome measures in early knee osteoarthritis Nat Rev Rheumatol 2019 15 438 448 31201386
  26. [26]Xu D, van der Voet J, Hansson M et al (2020) Association between meniscal volume and development of knee osteoarthritis. Rheumatology (Oxford). 10.1093/rheumatology/keaa522 PMC7937026 32974683 · DOI
  27. [27]Van der Lijn F De Bruijne M Klein S Automated brain structure segmentation based on atlas registration and appearance models IEEE Trans Med Imaging 2011 31 276 286 21937346
  28. [28]Klein S Staring M Murphy K Viergever MA Pluim JPW Elastix: a toolbox for intensity-based medical image registration IEEE Trans Med Imaging 2009 29 196 205 19923044
  29. [29]Burr DB Anatomy and physiology of the mineralized tissues: role in the pathogenesis of osteoarthrosis Osteoarthritis Cartilage 2004 12 20 30 14698637
  30. [30]Madry H van Dijk CN Mueller-Gerbl M The basic science of the subchondral bone Knee Surg Sports Traumatol Arthrosc 2010 18 419 433 20119671
  31. [32]Zou KH Warfield SK Bharatha A Statistical validation of image segmentation quality based on a spatial overlap index1: scientific reports Acad Radiol 2004 11 178 189 14974593 PMC1415224
  32. [33]Kerkhof HJM Bierma-Zeinstra SMA Arden NK Prediction model for knee osteoarthritis incidence, including clinical, genetic and biochemical risk factors Ann Rheum Dis 2014 73 2116 2121 23962456
  33. [34]Friedman J Hastie T Tibshirani R Regularization paths for generalized linear models via coordinate descent J Stat Softw 2010 33 1 22 20808728 PMC2929880
  34. [36]Nadeau C, Bengio Y (2000) Inference for the generalization error. Adv Neural Inf Process Syst, pp 307–313
  35. [38]Grau J Grosse I Keilwagen J PRROC: computing and visualizing precision-recall and receiver operating characteristic curves in R Bioinformatics 2015 31 2595 2597 25810428 PMC4514923
  36. [39]Saito T Rehmsmeier M Precrec: fast and accurate precision–recall and ROC curve calculations in R Bioinformatics 2017 33 145 147 27591081 PMC5408773
  37. [40]McAlindon TE Snow S Cooper C Dieppe PA Radiographic patterns of osteoarthritis of the knee joint in the community: the importance of the patellofemoral joint Ann Rheum Dis 1992 51 844 849 1632657 PMC1004766
  38. [41]Wise BL Niu J Yang M Patterns of compartment involvement in tibiofemoral osteoarthritis in men and women and in whites and African Americans Arthritis Care Res 2012 64 847 852 PMC3340516 22238208
  39. [42]Kraus VB Feng S Wang S Subchondral bone trabecular integrity predicts and changes concurrently with radiographic and magnetic resonance imaging–determined knee osteoarthritis progression Arthritis Rheum 2013 65 1812 1821 23576116 PMC4152231
  40. [43]Kraus VB Collins JE Charles HC Predictive validity of radiographic trabecular bone texture in knee osteoarthritis: the Osteoarthritis Research Society International/Foundation for the National Institutes of Health Osteoarthritis Biomarkers Consortium Arthritis Rheumatol 2018 70 80 87 29024470 PMC5745253
  41. [44]Janvier T Jennane R Toumi H Lespessailles E Subchondral tibial bone texture predicts the incidence of radiographic knee osteoarthritis: data from the osteoarthritis initiative Osteoarthritis Cartilage 2017 25 2047 2054 28935435
  42. [45]Hirvasniemi J Niinimäki J Thevenot J Saarakkala S Bone density and texture from minimally post-processed knee radiographs in subjects with knee osteoarthritis Ann Biomed Eng 2019 47 1181 1190 30767134 PMC6453872
  43. [46]Thomson J, O’Neill T, Felson D, Cootes T (2015) Automated shape and texture analysis for detection of osteoarthritis from radiographs of the knee. International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp 127–134
  44. [48]Grigorescu SE Petkov N Kruizinga P Comparison of texture features based on Gabor filters IEEE Trans Image Process 2002 11 1160 1167 18249688
  45. [50]Ambellan F Tack A Ehlke M Zachow S Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the Osteoarthritis Initiative Med Image Anal 2019 52 109 118 30529224

Authors
Hirvasniemi, J., Klein, S., Bierma-Zeinstra, S., Vernooij, M. W., Schiphof, D., & Oei, E. H. G.
Affiliations
Erasmus MC (NL)
Corresponding author
Oei, Edwin H. G.
ORCID
Venue
European Radiology
Publisher
Springer Science and Business Media LLC
Published
2021-04-21
Volume
31
Issue
11
Pages
8513-8521
ISSN
  • 0938-7994 (ISSN-L)
  • 1432-1084
DOI
10.1007/s00330-021-07951-5
PMC
PMC8523397
OpenAlex
W3155312544
Cited by
63
References
50
Citations by year
2026: 8 · 2025: 15 · 2024: 16 · 2023: 11 · 2022: 13
FWCI
5.28
Type
journal article
Language
en
Open access
Licence
Locations
  • European Radiology (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • EUR Research Repository (Erasmus University Rotterdam) (publishedVersion, cc-by)
  • EUR Research Repository (Erasmus University Rotterdam) (submittedVersion, cc-by)
  • RePub (Erasmus University Rotterdam) (submittedVersion, other-oa)
Funders
H2020 Marie Skłodowska-Curie Actions (707404)
Topics
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
  • Health Sciences › Medicine › Rheumatology › Osteoarthritis Treatment and Mechanisms
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Retrieval and Classification Techniques
Keywords
knee osteoarthritis, subchondral bone, tibia, MRI radiomics, imaging biomarkers, tibiofemoral osteoarthritis, MOAKS, multi-atlas segmentation, Rotterdam Study
MeSH terms
Machine Learning, Female, Humans, Knee Joint, Magnetic Resonance Imaging, ROC Curve, Tibia, Osteoarthritis, Knee

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review