Machine-learning, MRI bone shape and important clinical outcomes in osteoarthritis: data from the Osteoarthritis Initiative

Abstract

ObjectivesOsteoarthritis (OA) structural status is imperfectly classified using radiographic assessment. Statistical shape modelling (SSM), a form of machine-learning, provides precise quantification of a characteristic 3D OA bone shape. We aimed to determine the benefits of this novel measure of OA status for assessing risks of clinically important outcomes.MethodsThe study used 4796 individuals from the Osteoarthritis Initiative cohort. SSM-derived femur bone shape (B-score) was measured from all 9433 baseline knee MRIs. We examined the relationship between B-score, radiographic Kellgren-Lawrence grade (KLG) and current and future pain and function as well as total knee replacement (TKR) up to 8 years.ResultsB-score repeatability supported 40 discrete grades. KLG and B-score were both associated with risk of current and future pain, functional limitation and TKR; logistic regression curves were similar. However, each KLG included a wide range of B-scores. For example, for KLG3, risk of pain was 34.4 (95% CI 31.7 to 37.0)%, but B-scores within KLG3 knees ranged from 0 to 6; for B-score 0, risk was 17.0 (16.1 to 17.9)% while for B-score 6, it was 52.1 (48.8 to 55.4)%. For TKR, KLG3 risk was 15.3 (13.3 to 17.3)%; while B-score 0 had negligible risk, B-score 6 risk was 35.6 (31.8 to 39.6)%. Age, sex and body mass index had negligible effects on association between B-score and symptoms.ConclusionsB-score provides reader-independent quantification using a single time-point, providing unambiguous OA status with defined clinical risks across the whole range of disease including pre-radiographic OA. B-score heralds a step-change in OA stratification for interventions and improved personalised assessment, analogous to the T-score in osteoporosis.

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

“…Radiomic methods have been successfully applied to different MRI data [14] but these methods have not yet been widely used for the assessment of knee osteoarthritis. Bone shape and texture variables have been previously extracted from knee MRI to assess osteoarthritic changes [15][16][17][18] and texture has been shown to correlate with the actual three-dimensional microstructure of subchondral bone [19]. However, previous texture analysis studies have manually defined regions of interests, analyzed limited number of slices, and used a limited number of texture variables.…”

Study TPX-100-5: Intra-Articular TPX-100 Significantly Delays Pathological Bone Shape Change and Stabilizes Cartilage in Moderate to Severe Bilateral Knee OA

“…Distances along the femur OA vector are termed 'B-score', with the origin (B-score = 0) de ned as the mean shape of the Non-OA Group for each sex, and 1 unit de ned as 1 standard deviation of the Non-OA Group along the OA vector (positive values toward the OA Group). Each parameterized femur bone shape was projected orthogonally onto the OA vector to specify the corresponding B-score value [10]. Representative examples of differences in femur bone shape at various B-scores, and a heat map of the areas which change most with increasing B-score are shown in Fig.…”

“…The average slope over this 4-year period is about 0.24 B-scores per annum. SDD was calculated as the 95% limit of agreement between the rst and second image measurements from test-retest data, using the Bland-Altman method [10].…”

“…These data provide construct validity for this new, continuous scalar measurement. In addition, Bowes et al showed that bone shape is directly associated with the risk of clinical outcome measures such as knee pain, functional de cit and joint failure, as indicated by total joint replacement, with only small effect sizes from adjusting models for potential covariates such as age, sex, ethnicity, body mass index (BMI), alignment, previous knee surgery, non-steroidal anti-in ammatory drugs (NSAIDs) use and smoking status [10]. These ndings support femoral B-score as a structural endpoint in clinical trials of diseasemodifying osteoarthritis drugs (DMOADs).…”

“…The femoral bone shape ("B-score") metric is a form of statistical z-score that represents the position of a femoral bone shape along a shape vector from a non-OA knee shape (origin) towards an OA knee shape (positive direction). Non-OA and OA knees used to de ne this 3-dimensional (3D) shape vector were categorized using centrally-read and adjudicated Kellgren-Lawrence grading [10].…”