Deep learning approach to predict pain progression in knee osteoarthritis

Guan, Bochen; Liu, F.; Matthew, P Abdel; Mirzaian, Arya; Demehri, Shadpour; Neogi, Tuhina; Guermazi, Ali; Kijowski, Richard

doi:10.1016/j.joca.2020.02.489

Cited by 7 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, also OA research has increasingly started focusing on the development of different approaches and methods based on machine learning (ML) algorithms [4]- [7] and nite element analysis (FEA) [8]- [10] to classify subjects at high risk for knee OA development. Especially, the aim has been to identify the high-risk subjects before any degenerative signs are detected from a clinical image.…”

Section: Introductionmentioning

confidence: 99%

“…Especially, the aim has been to identify the high-risk subjects before any degenerative signs are detected from a clinical image. In ML approaches [4]- [7], the prediction is based on the parameters that can be easily quanti ed, such as age, height, weight, parameters that are evaluated from clinical image, such as Kellgren-Lawrence (KL) grade or tibiofemoral angle, and parameters that are reported by the subject itself such as different physical activity levels and pain indexes. This generates a huge set of parameters that needs to be de ned before making an accurate prediction.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optimized simplified constitutive models with an atlas-based finite element analysis can be utilized to predict personalized progression of knee osteoarthritis: Data from the Osteoarthritis Initiative

Mononen

Carvajal

Liukkonen

et al. 2023

Preprint

View full text Add to dashboard Cite

New technologies are required to support a radical shift towards preventive healthcare. Here we focus on evaluating the possibility of finite element (FE) analysis-aided prevention of knee osteoarthritis (OA), a disease that affects 100 million citizens in the US and EU and this number is estimated to increase drastically. Current clinical methods to diagnose or predict joint health status relies on symptoms and tissue failures obtained from clinical imaging. In a joint with no detectable injuries, the diagnosis of the future health of the knee can be assumed to be very subjective. Quantitative approaches are therefore needed to assess the personalized risk for the onset and development of knee OA. FE analysis utilizing an atlas-based modeling approach has shown a preliminary capability for simulating subject-specific cartilage mechanical responses. However, it has been verified with a very limited subject number. Thus, the aim of this study is to verify the real capability of the atlas-based approach to simulate cartilage degeneration utilizing different material descriptions for cartilage. A fibril reinforced poroviscoelastic (FRPVE) material formulation was considered as state-of-the-art material behavior and simulated mechanical tissue responses and predicted cartilage degenerations within knee joint were compared against to simpler constitutive models for cartilage. The capability of the atlas-based modeling to offer a feasible approach with quantitative evaluation for the risk for the OA development (healthy vs osteoarthritic knee, p < 0.01, AUC ~ 0.7) was verified with 214 knees. Furthermore, the results suggest that accuracy for simulation of cartilage degeneration with simpler material models is similar to models using FPRVE materials if the material parameters are chosen properly.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Optimized simplified constitutive models with an atlas-based finite element analysis can be utilized to predict personalized progression of knee osteoarthritis: Data from the Osteoarthritis Initiative

Mononen

Carvajal

Liukkonen

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The authors used a combined joint training model, first utilizing a deep learning network to extract information from baseline knee radiographs as a feature vector, followed by concatenation with risk factor data vector [65]. Guan et al [66] also recently published a study using radiographs (deep learning artificial neural networks) combined with demographic data (i.e., BMI, gender) to predict knee pain progression (defined as 9-point or greater increase in WOMAC score (0-100 scale) between 2 or more time points from 24-month to 60-month follow-up) yielding an AUC of 0.80 with 75.2% sensitivity and 76.2% specificity. Their ensemble model was created by integrating both clinical data and deep learning analysis of the baseline knee radiographs.…”

Section: Radiograph-based Ensemble ML Modelsmentioning

confidence: 99%

AI MSK clinical applications: cartilage and osteoarthritis

et al. 2021

View full text Add to dashboard Cite

The advancements of artificial intelligence (AI) for osteoarthritis (OA) applications have been rapid in recent years, particularly innovations of deep learning for image classification, lesion detection, cartilage segmentation, and prediction modeling of future knee OA development. This review article focuses on AI applications in OA research, first describing machine learning (ML) techniques and workflow, followed by how these algorithms are used for OA classification tasks through imaging and non-imaging-based ML models. Deep learning applications for OA research, including analysis of both radiographs for automatic detection of OA severity, and MR images for detection of cartilage/meniscus lesions and cartilage segmentation for automatic T 2 quantification will be described. In addition, information on ML models that identify individuals at high risk of OA development will be provided. The future vision of machine learning applications in imaging of OA and cartilage hinges on implementation of AI for optimizing imaging protocols, quantitative assessment of cartilage, and automated analysis of disease burden yielding a faster and more efficient workflow for a radiologist with a higher level of reproducibility and precision. It may also provide risk assessment tools for individual patients, which is an integral part of precision medicine.

show abstract

Associating Knee Osteoarthritis Progression with Temporal‐Regional Graph Convolutional Network Analysis on MR Images

Hu,

Peng,

Zhou

et al. 2024

Magnetic Resonance Imaging

View full text Add to dashboard Cite

BackgroundArtificial intelligence shows promise in assessing knee osteoarthritis (OA) progression on MR images, but faces challenges in accuracy and interpretability.PurposeTo introduce a temporal‐regional graph convolutional network (TRGCN) on MR images to study the association between knee OA progression status and network outcome.Study TypeRetrospective.Population194 OA progressors (mean age, 62 ± 9 years) and 406 controls (mean age, 61 ± 9 years) from the OA Initiative were randomly divided into training (80%) and testing (20%) cohorts.Field Strength/SequenceSagittal 2D IW‐TSE‐FS (IW) and 3D‐DESS‐WE (DESS) at 3T.AssessmentAnatomical subregions of cartilage, subchondral bone, meniscus, and the infrapatellar fat pad at baseline, 12‐month, and 24‐month were automatically segmented and served as inputs to form compartment‐based graphs for a TRGCN model, which containing both regional and temporal information. The performance of models based on (i) clinical variables alone, (ii) radiologist score alone, (iii) combined features (containing i and ii), (iv) composite TRGCN (combining TRGCN, i and ii), (v) radiomics features, (vi) convolutional neural network based on Densenet‐169 were compared.Statistical TestsDeLong test was performed to compare the areas under the ROC curve (AUC) of all models. Additionally, interpretability analysis was done to evaluate the contributions of individual regions. A P value <0.05 was considered significant.ResultsThe composite TRGCN outperformed all other models with AUCs of 0.841 (DESS) and 0.856 (IW) in the testing cohort (all P < 0.05). Interpretability analysis highlighted cartilage's importance over other structures (42%–45%), tibiofemoral joint's (TFJ) dominance over patellofemoral joint (PFJ) (58%–67% vs. 12%–37%), and importance scores changes in compartments over time (TFJ vs. PFJ: baseline: 44% vs. 43%, 12‐month: 52% vs. 39%, 24‐month: 31% vs. 48%).Data ConclusionThe composite TRGCN, capturing temporal and regional information, demonstrated superior discriminative ability compared with other methods, providing interpretable insights for identifying knee OA progression.Level of Evidence4.Technical EfficacyStage 2.

show abstract

Deep learning approach to predict pain progression in knee osteoarthritis

Cited by 7 publications

References 0 publications

Optimized simplified constitutive models with an atlas-based finite element analysis can be utilized to predict personalized progression of knee osteoarthritis: Data from the Osteoarthritis Initiative

Optimized simplified constitutive models with an atlas-based finite element analysis can be utilized to predict personalized progression of knee osteoarthritis: Data from the Osteoarthritis Initiative

AI MSK clinical applications: cartilage and osteoarthritis

Associating Knee Osteoarthritis Progression with Temporal‐Regional Graph Convolutional Network Analysis on MR Images

Contact Info

Product

Resources

About