Combining Machine-Learning Assessment of Multiple MRI Pathologies and Clinical Phenotypes for Predicting Joint Replacement in Knee Osteoarthritis: Data From the Osteoarthritis Initiative.
ObjectiveArtificial intelligence offers opportunities for timesaving assessments of multiple pathologies in large magnetic resonance imaging (MRI) data sets in knee osteoarthritis (KOA). This study evaluated their prevalence within pre-defined clinical phenotypes and their predictive value for knee replacement (KR).DesignBaseline MRIs (n = 8,667) from the Osteoarthritis Initiative were analyzed using a machine-learning (ML) algorithm. The presence of pathologies (menisci, anterior cruciate, medial collateral ligaments, cartilage, etc.) was assessed in previously identified phenotypic clusters (a post-traumatic, metabolic, and age-defined phenotype). The value of both, cluster allocation and joint pathology for KR prediction was evaluated using supervised ML models and time-dependent receiver operating characteristic curves.ResultsCompared to the population average, the metabolic cluster had a higher prevalence of cartilage lesions, while the post-traumatic one had more medial meniscal damage. Random forest models showed the best prediction (area under the curve 0.837, test set at 2 years). The top predictors for KR were meniscal position (relative to the border of the tibial plateau), severe joint effusion, medial femorotibial cartilage lesions, and metabolic phenotype. These features defined patients at high risk of KR with an estimated KR rate at 5 years of 10% vs 3% in the high- and low-risk groups based on a predictive risk score including all analyzed structures.ConclusionsThis ML-enabled assessment of multiple MRI pathologies in a large KOA data set highlights the importance of meniscal pathologies and markers of inflammation, in addition to cartilage assessments and clinical information for patient stratification and improved prediction of KOA progression to KR.
- Research Article
20
- 10.1002/acr.24235
- Oct 1, 2020
- Arthritis Care & Research
Measures of Adult Knee Function.
- Research Article
- 10.2519/jospt.2010.40.1.a12
- Jan 1, 2010
- Journal of Orthopaedic & Sports Physical Therapy
CSM 2010 Orthopaedic Section Platform Presentations (Abstracts OPL1-OPL67)
- Front Matter
2
- 10.2106/jbjs.20.01406
- Sep 17, 2020
- The Journal of bone and joint surgery. American volume
What's New in Orthopaedic Rehabilitation.
- Front Matter
2
- 10.2106/jbjs.20.01753
- Dec 3, 2020
- Journal of Bone and Joint Surgery
Update This article was updated on February 6, 2019, because of a previous error. On page 105, in the subsection titled “Outcomes and Design” the sentence that had read “Furthermore, in a retrospective review, Houdek et al. 48 , at a mean follow-up of 8 years, demonstrated improved survivorship of 9,999 metal-backed compared with 1,645 all-polyethylene tibial components, over all age groups and most BMI categories” now reads “Furthermore, in a retrospective review, Houdek et al. 48 , at a mean follow-up of 8 years, demonstrated inferior survivorship of 9,999 metal-backed compared with 1,645 all-polyethylene tibial components, over all age groups and most BMI categories.” An erratum has been published: J Bone Joint Surg Am. 2019 Mar 20;101(6):e26.
- Abstract
- 10.1016/j.joca.2021.02.088
- Apr 1, 2021
- Osteoarthritis and Cartilage
Automated morphological phenotyping to characterize knee osteoarthritis progression
- Research Article
- 10.1016/j.joca.2018.02.800
- Apr 1, 2018
- Osteoarthritis and Cartilage
3D meniscal position predicts knee replacement in fast progressing knee osteoarthritis
- Research Article
1
- 10.2106/jbjs.22.01030
- Nov 16, 2022
- Journal of Bone and Joint Surgery
What's New in Adult Reconstructive Knee Surgery.
- Research Article
- 10.2519/jospt.2012.42.1.a1
- Jan 1, 2012
- Journal of Orthopaedic & Sports Physical Therapy
CSM 2012 Orthopaedic and Sports Physical Therapy Section Programming
- Abstract
- 10.1016/j.joca.2014.02.219
- Mar 20, 2014
- Osteoarthritis and Cartilage
Quantifying differences in passive knee laxity and finite helical axis measures between healthy and anterior cruciate ligament deficient individuals
- Abstract
- 10.1016/j.joca.2016.01.094
- Mar 20, 2016
- Osteoarthritis and Cartilage
The value of total knee replacement revisited: A comparative analysis of its effectiveness and impact on healthcare costs using data from the osteoarthritis initiative
- Research Article
90
- 10.1002/jor.1100170107
- Jan 1, 1999
- Journal of Orthopaedic Research
Early expression of marker genes in the rabbit medial collateral and anterior cruciate ligaments: The use of different viral vectors and the effects of injury
- Front Matter
22
- 10.1302/0301-620x.99b2.38085
- Feb 1, 2017
- The Bone & Joint Journal
In this issue of The BJJ, we have two specialty updates in knee surgery that will be of interest to clinicians all around the world. Jonathan Eldridge and his team provide a clear perspective on adolescent patellar instability – a problem with many variations and even more solutions.[1][1] Wang
- Abstract
- 10.1177/2325967121s00803
- Jul 1, 2022
- Orthopaedic Journal of Sports Medicine
Poster 242: Deep Learning for Identifying Patellofemoral Measurements Associated with Cartilage Lesions on MRI
- Research Article
26
- 10.1111/os.12395
- Aug 1, 2018
- Orthopaedic Surgery
Knee osteoarthritis (OA) is a prevalent disease in the elderly, causing pain and contributing to poor quality of life. Surgical intervention, such as knee arthroplasty, can be used in those with end-stage knee OA. Total knee arthroplasty (TKA) is one of the most common surgical procedures for end-stage knee OA, with promising clinical outcomes. However, a large proportion of patients with isolated compartment OA can be treated with unicompartmental knee arthroplasty (UKA) instead. UKA has shown better patient-reported functional outcomes, and lower mortality and major complication rates than TKA. The percentage of UKA in knee arthroplasty varied in different orthopedic centers, and we believed that the requirement for UKA was underestimated in many centers. A retrospective study was carried out on our Chinese patient population presenting for knee arthroplasty; it aimed to identify the proportion of patients that might be suitable for UKA. A retrospective cross-sectional study of 155 consecutive patients (168 knees) awaiting TKA for end-stage primary OA was performed. The pattern and grade of OA was recorded from preoperative weight-bearing anteroposterior and non-weight-bearing lateral radiographs. The medial, lateral, patellofemoral compartment was given an individual Kellgren-Lawrence grade on the radiographs, and those grade ≥3 were defined as end-stage OA. The compartments involvement was established then. The integrity of the anterior cruciate ligament (ACL) was determined by the modified Keyes classification on lateral radiographs. The applicability for total or partial knee arthroplasty was determined according to the compartments involvement. Medial compartment involvement was found in 154 (91.7%) knees, while the involvement of the lateral compartment and patellofemoral joint was found in 54 (32.1%) and 57 (33.9%) knees, respectively. Eighty-one (48.2%) of the knees showed medial compartment OA with or without patellofemoral joint involvement, and modified Keyes classification grade 1, indicating an intact ACL, and, hence, potential suitability for medial UKA. Isolated lateral OA indicating possible suitability for lateral UKA was identified in 11 knees (6.5%). No patients showed isolated patellofemoral joint OA. The other 76 (45.2%) knees could be treated by TKA. The medial compartment was the most commonly affected in our Chinese patients indicated for knee arthroplasty. More than half of the patients in this group could be treated by either medial or lateral UKA.
- Abstract
- 10.1136/annrheumdis-2016-eular.5728
- Jun 1, 2016
- Annals of the Rheumatic Diseases
BackgroundKnee osteoarthritis (OA) may supervene on traumatic anterior cruciate ligament (ACL) injury/rupture. However, knee OA with advanced disease on MRI may have ACL deficiency without a history of injury. We...
- Ask R Discovery
- Chat PDF
AI summaries and top papers from 250M+ research sources.