Related Experiment Video
Updated: Nov 30, 2025
07:32
Standardized Histomorphometric Evaluation of Osteoarthritis in a Surgical Mouse Model
Published on: May 6, 2020
12.8K
Machine-learning, MRI bone shape and important clinical outcomes in osteoarthritis: data from the Osteoarthritis
Michael A Bowes1, Katherine Kacena2, Oras A Alabas3
1Imorphics Ltd, Manchester, UK.
Annals of the Rheumatic Diseases
|November 14, 2020
Summary
Statistical shape modelling (SSM) offers a precise measure of osteoarthritis (OA) bone shape, improving risk assessment for pain, function, and knee replacement compared to traditional grading. This novel B-score provides clearer OA status for personalized treatment.
Area of Science:
- Orthopedics and Sports Medicine
- Medical Imaging
- Machine Learning in Healthcare
Background:
- Radiographic assessment of osteoarthritis (OA) structural status is often imprecise.
- Statistical shape modelling (SSM) offers a machine-learning approach for precise quantification of 3D bone shape in OA.
Purpose of the Study:
- To evaluate the benefits of SSM-derived bone shape (B-score) for assessing OA status and predicting clinical outcomes.
- To compare the predictive value of B-score against traditional radiographic Kellgren-Lawrence grade (KLG).
Main Methods:
- Utilized data from 4796 individuals in the Osteoarthritis Initiative cohort.
- Measured SSM-derived femur bone shape (B-score) from 9433 baseline knee MRIs.
- Examined associations between B-score, KLG, and risks of pain, function, and total knee replacement (TKR) over 8 years.
Main Results:
- B-score demonstrated good repeatability, supporting 40 discrete grades.
- Both B-score and KLG correlated with risks of current/future pain, functional limitation, and TKR.
- B-score provided more granular risk stratification within KLG grades, particularly for severe OA (KLG3), showing significantly different risks for low (B-score 0) versus high (B-score 6) shape deviation.
Conclusions:
- B-score offers reader-independent, precise OA quantification from a single time-point, defining clinical risks across all disease stages, including pre-radiographic OA.
- B-score represents a significant advancement in OA stratification for interventions and personalized patient assessment, similar to the T-score in osteoporosis.
