1
|
Hirvasniemi J, Runhaar J, van der Heijden RA, Zokaeinikoo M, Yang M, Li X, Tan J, Rajamohan HR, Zhou Y, Deniz CM, Caliva F, Iriondo C, Lee JJ, Liu F, Martinez AM, Namiri N, Pedoia V, Panfilov E, Bayramoglu N, Nguyen HH, Nieminen MT, Saarakkala S, Tiulpin A, Lin E, Li A, Li V, Dam EB, Chaudhari AS, Kijowski R, Bierma-Zeinstra S, Oei EHG, Klein S. The KNee OsteoArthritis Prediction (KNOAP2020) challenge: An image analysis challenge to predict incident symptomatic radiographic knee osteoarthritis from MRI and X-ray images. Osteoarthritis Cartilage 2023; 31:115-125. [PMID: 36243308 DOI: 10.1016/j.joca.2022.10.001] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/01/2022] [Revised: 09/02/2022] [Accepted: 10/03/2022] [Indexed: 11/05/2022]
Abstract
OBJECTIVES The KNee OsteoArthritis Prediction (KNOAP2020) challenge was organized to objectively compare methods for the prediction of incident symptomatic radiographic knee osteoarthritis within 78 months on a test set with blinded ground truth. DESIGN The challenge participants were free to use any available data sources to train their models. A test set of 423 knees from the Prevention of Knee Osteoarthritis in Overweight Females (PROOF) study consisting of magnetic resonance imaging (MRI) and X-ray image data along with clinical risk factors at baseline was made available to all challenge participants. The ground truth outcomes, i.e., which knees developed incident symptomatic radiographic knee osteoarthritis (according to the combined ACR criteria) within 78 months, were not provided to the participants. To assess the performance of the submitted models, we used the area under the receiver operating characteristic curve (ROCAUC) and balanced accuracy (BACC). RESULTS Seven teams submitted 23 entries in total. A majority of the algorithms were trained on data from the Osteoarthritis Initiative. The model with the highest ROCAUC (0.64 (95% confidence interval (CI): 0.57-0.70)) used deep learning to extract information from X-ray images combined with clinical variables. The model with the highest BACC (0.59 (95% CI: 0.52-0.65)) ensembled three different models that used automatically extracted X-ray and MRI features along with clinical variables. CONCLUSION The KNOAP2020 challenge established a benchmark for predicting incident symptomatic radiographic knee osteoarthritis. Accurate prediction of incident symptomatic radiographic knee osteoarthritis is a complex and still unsolved problem requiring additional investigation.
Collapse
Affiliation(s)
- J Hirvasniemi
- Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands.
| | - J Runhaar
- Department of General Practice, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| | - R A van der Heijden
- Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| | - M Zokaeinikoo
- Department of Biomedical Engineering, Cleveland Clinic, Cleveland, USA
| | - M Yang
- Department of Biomedical Engineering, Cleveland Clinic, Cleveland, USA
| | - X Li
- Department of Biomedical Engineering, Cleveland Clinic, Cleveland, USA
| | - J Tan
- Department of Radiology, New York University Langone Health, New York, USA
| | - H R Rajamohan
- Department of Radiology, New York University Langone Health, New York, USA
| | - Y Zhou
- Department of Radiology, New York University Langone Health, New York, USA
| | - C M Deniz
- Department of Radiology, New York University Langone Health, New York, USA
| | - F Caliva
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - C Iriondo
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - J J Lee
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - F Liu
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - A M Martinez
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - N Namiri
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - V Pedoia
- Department of Radiology, University of California, San Francisco, San Francisco, USA
| | - E Panfilov
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
| | - N Bayramoglu
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
| | - H H Nguyen
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
| | - M T Nieminen
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
| | - S Saarakkala
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland; Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland
| | - A Tiulpin
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
| | - E Lin
- Akousist Co., Ltd., Taoyuan City, Taiwan
| | - A Li
- Akousist Co., Ltd., Taoyuan City, Taiwan
| | - V Li
- Akousist Co., Ltd., Taoyuan City, Taiwan
| | - E B Dam
- Department of Computer Science, University of Copenhagen, Copenhagen, Denmark
| | - A S Chaudhari
- Department of Radiology, Stanford University, Stanford, USA
| | - R Kijowski
- Department of Radiology, New York University Langone Health, New York, USA
| | - S Bierma-Zeinstra
- Department of General Practice, Erasmus MC University Medical Center, Rotterdam, the Netherlands; Department of Orthopedics & Sport Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| | - E H G Oei
- Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| | - S Klein
- Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| |
Collapse
|
2
|
Gebre RK, Hirvasniemi J, van der Heijden RA, Lantto I, Saarakkala S, Leppilahti J, Jämsä T. Detecting hip osteoarthritis on clinical CT: a deep learning application based on 2-D summation images derived from CT. Osteoporos Int 2022; 33:355-365. [PMID: 34476540 PMCID: PMC8813821 DOI: 10.1007/s00198-021-06130-y] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/23/2020] [Accepted: 08/20/2021] [Indexed: 10/27/2022]
Abstract
UNLABELLED We developed and compared deep learning models to detect hip osteoarthritis on clinical CT. The CT-based summation images, CT-AP, that resemble X-ray radiographs can detect radiographic hip osteoarthritis and in the absence of large training data, a reliable deep learning model can be optimized by combining CT-AP and X-ray images. INTRODUCTION In this study, we aimed to investigate the applicability of deep learning (DL) to assess radiographic hip osteoarthritis (rHOA) on computed tomography (CT). METHODS The study data consisted of 94 abdominopelvic clinical CTs and 5659 hip X-ray images collected from Cohort Hip and Cohort Knee (CHECK). The CT slices were sequentially summed to create radiograph-like 2-D images named CT-AP. X-ray and CT-AP images were classified as rHOA if they had osteoarthritic changes corresponding to Kellgren-Lawrence grade 2 or higher. The study data was split into 55% training, 30% validation, and 15% test sets. A pretrained ResNet18 was optimized for a classification task of rHOA vs. no-rHOA. Five models were trained using (1) X-rays, (2) downsampled X-rays, (3) combination of CT-AP and X-ray images, (4) combination of CT-AP and downsampled X-ray images, and (5) CT-AP images. RESULTS Amongst the five models, Model-3 and Model-5 performed best in detecting rHOA from the CT-AP images. Model-3 detected rHOA on the test set of CT-AP images with a balanced accuracy of 82.2% and was able to discriminate rHOA from no-rHOA with an area under the receiver operating characteristic curve (ROC AUC) of 0.93 [0.75-0.99]. Model-5 detected rHOA on the test set at a balanced accuracy of 82.2% and classified rHOA from no-rHOA with an ROC AUC of 0.89 [0.67-0.97]. CONCLUSION CT-based summation images that resemble radiographs can be used to detect rHOA. In addition, in the absence of large training data, a reliable DL model can be optimized by combining CT-AP and X-ray images.
Collapse
Affiliation(s)
- R K Gebre
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland.
| | - J Hirvasniemi
- Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| | - R A van der Heijden
- Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
| | - I Lantto
- Division of Orthopaedic and Trauma Surgery, Oulu University Hospital, Oulu, Finland
- Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland
| | - S Saarakkala
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
- Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland
- Diagnostic Radiology, Oulu University Hospital, Oulu, Finland
| | - J Leppilahti
- Division of Orthopaedic and Trauma Surgery, Oulu University Hospital, Oulu, Finland
- Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland
| | - T Jämsä
- Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland
- Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland
- Diagnostic Radiology, Oulu University Hospital, Oulu, Finland
| |
Collapse
|
3
|
Oei EHG, Hirvasniemi J, van Zadelhoff TA, van der Heijden RA. Osteoarthritis year in review 2021: imaging. Osteoarthritis Cartilage 2022; 30:226-236. [PMID: 34838670 DOI: 10.1016/j.joca.2021.11.012] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/29/2021] [Revised: 10/16/2021] [Accepted: 11/11/2021] [Indexed: 02/02/2023]
Abstract
PURPOSE To provide a narrative review of original articles on imaging of osteoarthritis (OA) published between January 1, 2020 and March 31, 2021, with a special focus on imaging of inflammation, imaging of bone, cartilage and bone-cartilage interactions, imaging of peri-articular tissues, imaging scoring methods for OA, and artificial intelligence (AI) applied to OA imaging. METHODS The Embase, Pubmed, Medline, Cochrane databases were searched for original research articles in the English language on human, in vivo, imaging of OA published between January 1, 2020 and March 31, 2021. Search terms related to osteoarthritis combined with all imaging modalities and artificial intelligence were applied. A selection of articles reporting on one of the focus topics was discussed further. RESULTS The search resulted in 651 articles, of which 214 were deemed relevant to human OA imaging. Among the articles included, the knee joint (69%) and magnetic resonance imaging (MRI) (52%) were the predominant anatomical area and imaging modality studied. There were also a substantial number of papers (n = 46) reporting on AI applications in the field of OA imaging. CONCLUSION Imaging continues to play an important role in the assessment of OA. Recent advances in OA imaging include quantitative, non-contrast, and hybrid imaging techniques for improved characterization of multiple tissue processes in OA. In addition, an increasing effort in AI techniques is undertaken to enhance OA imaging acquisition and analysis.
Collapse
Affiliation(s)
- E H G Oei
- Department of Radiology & Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, Dr. Molewaterplein 40, 3015 GD Rotterdam, the Netherlands.
| | - J Hirvasniemi
- Department of Radiology & Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, Dr. Molewaterplein 40, 3015 GD Rotterdam, the Netherlands.
| | - T A van Zadelhoff
- Department of Radiology & Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, Dr. Molewaterplein 40, 3015 GD Rotterdam, the Netherlands.
| | - R A van der Heijden
- Department of Radiology & Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, Dr. Molewaterplein 40, 3015 GD Rotterdam, the Netherlands.
| |
Collapse
|
4
|
Kozijn AE, Gierman LM, van der Ham F, Mulder P, Morrison MC, Kühnast S, van der Heijden RA, Stavro PM, van Koppen A, Pieterman EJ, van den Hoek AM, Kleemann R, Princen HMG, Mastbergen SC, Lafeber FPJG, Zuurmond AM, Bobeldijk I, Weinans H, Stoop R. Variable cartilage degradation in mice with diet-induced metabolic dysfunction: food for thought. Osteoarthritis Cartilage 2018; 26:95-107. [PMID: 29074298 DOI: 10.1016/j.joca.2017.10.010] [Citation(s) in RCA: 22] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/01/2017] [Revised: 10/03/2017] [Accepted: 10/10/2017] [Indexed: 02/02/2023]
Abstract
OBJECTIVE Human cohort studies have demonstrated a role for systemic metabolic dysfunction in osteoarthritis (OA) pathogenesis in obese patients. To explore the mechanisms underlying this metabolic phenotype of OA, we examined cartilage degradation in the knees of mice from different genetic backgrounds in which a metabolic phenotype was established by various dietary approaches. DESIGN Wild-type C57BL/6J mice and genetically modified mice (hCRP, LDLr-/-. Leiden and ApoE*3Leiden.CETP mice) based on C57BL/6J background were used to investigate the contribution of inflammation and altered lipoprotein handling on diet-induced cartilage degradation. High-caloric diets of different macronutrient composition (i.e., high-carbohydrate or high-fat) were given in regimens of varying duration to induce a metabolic phenotype with aggravated cartilage degradation relative to controls. RESULTS Metabolic phenotypes were confirmed in all studies as mice developed obesity, hypercholesteremia, glucose intolerance and/or insulin resistance. Aggravated cartilage degradation was only observed in two out of the twelve experimental setups, specifically in long-term studies in male hCRP and female ApoE*3Leiden.CETP mice. C57BL/6J and LDLr-/-. Leiden mice did not develop HFD-induced OA under the conditions studied. Osteophyte formation and synovitis scores showed variable results between studies, but also between strains and gender. CONCLUSIONS Long-term feeding of high-caloric diets consistently induced a metabolic phenotype in various C57BL/6J (-based) mouse strains. In contrast, the induction of articular cartilage degradation proved variable, which suggests that an additional trigger might be necessary to accelerate diet-induced OA progression. Gender and genetic modifications that result in a humanized pro-inflammatory state (human CRP) or lipoprotein metabolism (human-E3L.CETP) were identified as important contributing factors.
Collapse
Affiliation(s)
- A E Kozijn
- Metabolic Health Research, TNO, Leiden, The Netherlands; Department of Orthopaedics, University Medical Center (UMC) Utrecht, Utrecht University, Utrecht, The Netherlands; Department of Rheumatology & Clinical Immunology, UMC Utrecht, Utrecht University, Utrecht, The Netherlands
| | - L M Gierman
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - F van der Ham
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - P Mulder
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - M C Morrison
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - S Kühnast
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - R A van der Heijden
- Department of Pathology and Medical Biology, UMC Groningen, Groningen, The Netherlands
| | - P M Stavro
- Bunge North America, Saint Louis, United States
| | - A van Koppen
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - E J Pieterman
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | | | - R Kleemann
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - H M G Princen
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - S C Mastbergen
- Department of Rheumatology & Clinical Immunology, UMC Utrecht, Utrecht University, Utrecht, The Netherlands
| | - F P J G Lafeber
- Department of Rheumatology & Clinical Immunology, UMC Utrecht, Utrecht University, Utrecht, The Netherlands
| | - A-M Zuurmond
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - I Bobeldijk
- Metabolic Health Research, TNO, Leiden, The Netherlands
| | - H Weinans
- Department of Orthopaedics, University Medical Center (UMC) Utrecht, Utrecht University, Utrecht, The Netherlands; Department of Rheumatology & Clinical Immunology, UMC Utrecht, Utrecht University, Utrecht, The Netherlands; Department of Biomechanical Engineering, Delft University of Technology, Delft, The Netherlands
| | - R Stoop
- Metabolic Health Research, TNO, Leiden, The Netherlands.
| |
Collapse
|