Subchondral Bone Length in Knee Osteoarthritis: A Deep Learning–Derived Imaging Measure and Its Association With Radiographic and Clinical Outcomes

Abstract

Objective. To develop a bone shape measure that reflects the extent of cartilage loss and bone flattening in knee osteoarthritis (OA) and test it against estimates of disease severity.Methods. A fast region-based convolutional neural network was trained to crop the knee joints in sagittal dualecho steady-state magnetic resonance imaging sequences obtained from the Osteoarthritis Initiative (OAI). Publicly available annotations of the cartilage and menisci were used as references to annotate the tibia and the femur in 61 knees. Another deep neural network (U-Net) was developed to learn these annotations. Model predictions were compared to radiologist-driven annotations on an independent test set (27 knees). The U-Net was applied to automatically extract the knee joint structures on the larger OAI data set (n = 9,434 knees). We defined subchondral bone length (SBL), a novel shape measure characterizing the extent of overlying cartilage and bone flattening, and examined its relationship with radiographic joint space narrowing (JSN), concurrent pain and disability (according to the Western Ontario and McMaster Universities Osteoarthritis Index), as well as subsequent partial or total knee replacement. Odds ratios (ORs) and 95% confidence intervals (95% CIs) for each outcome were estimated using relative changes in SBL from the OAI data set stratified into quartiles.Results. The mean SBL values for knees with JSN were consistently different from knees without JSN. Greater changes of SBL from baseline were associated with greater pain and disability. For knees with medial or lateral JSN, the ORs for future knee replacement between the lowest and highest quartiles corresponding to SBL changes were 5.68 (95% CI 3.90-8.27) and 7.19 (95% CI 3.71-13.95), respectively.Conclusion. SBL quantified OA status based on JSN severity and shows promise as an imaging marker in predicting clinical and structural OA outcomes.

Performance of Radiological and Biochemical Biomarkers in Predicting Radio-Symptomatic Knee Osteoarthritis Progression

“…In the current study, the included imaging-based biomarkers were limited to radiography. Another limitation is the lack of investigation of the association between radiographic KOA progression and 3D MRI bone texture [ 34 ] or shape [ 35 ], or other MRI-based features such as bone marrow lesions [ 36 ]. Integrating such parameters into our model is worth assessing.…”

The application of machine learning in early diagnosis of osteoarthritis: a narrative review

“…However, due to the limited sample size and selection bias, the results can only be partially applied to patients at the early stages of ROA. 34 Chang et al 35 used a methodology based on bone flattening to estimate the presence of osteophytes and ROA (KL ⩾ 2). The AUC for osteophytes was 0.85.…”

Magnetic resonance imaging assessments for knee segmentation and their use in combination with machine/deep learning as predictors of early osteoarthritis diagnosis and prognosis

“… 9 , 121 , 142 – 144 Other methodologies included the bone shape vector 145 and the subchondral bone length. 146 Yet, on the one hand, the bone shape vector was developed only for one bone, the femur, and included the osteophytes in its measurement, which may induce inaccuracy in bone shape measurement changes. On the other hand, the subchondral bone length segmentation used U-Net and 2D shape measurement, which characterizes the degree of overlying bone flattening.…”