Skip to content

Publication Category: Original Article

Clinical Validation of Artificial Intelligence (AI)-based Cartilage Segmentation Predicting Knee Replacement

Objective: For cartilage morphology to serve as a scalable endpoint in clinical trials, analyses should be performed automatically without human interaction. To clinically validate artificial intelligence (AI)-based analysis, we studied cartilage loss from MRI prior to knee replacement. Methods: Knee replacement cases between 36 and 60 months’ follow-up in the Osteoarthritis Initiative were matched with controls by … Continued

Sex-related differences in femorotibial cartilage thickness: A matched pair analysis

Background: Women are more frequently affected by knee osteoarthritis (OA) than men. Differences in knee cartilage thickness have been proposed as an explanation, but it remains unclear whether these differences reflect true sex-specific variation or differences in body size. Clarifying this may improve our understanding of sex-specific mechanisms in knee OA development. The objective of this … Continued

Exploratory analysis of the 2-year changes in knee cartilage thickness and transverse relaxation time (T2) in ACL-injured versus healthy participants

Objective: To investigate the 2-year changes in cartilage thickness and transverse relaxation time (T2) in magnetic resonance images (MRI) of ACL-injured (ACL_in) and uninjured contralateral knees (ACL_unin), compared to healthy controls (HEA). Method: Baseline and 2-year follow-up MRIs were acquired in both knees of 78 participants (ACL-injured 2-10 years prior, 20-30 years, n = 20; 40-60 years, … Continued

Magnetic resonance imaging biomarkers of knee osteoarthritis progression

Objective: The Foundation for the National Institutes of Health (FNIH) OA Biomarkers Consortium aims to identify, develop, and qualify biomarkers to support drug development in knee osteoarthritis (OA). The project’s second phase, the PROGRESS OA study, aims to externally validate prognostic and response biomarkers identified in the earlier phase (phase 1). Here we present results assessing … Continued

Fluctuation of bone marrow lesions and inflammatory MRI markers over 2 years and concurrent associations with quantitative cartilage loss

ObjectiveTo assess whether change of semiquantitatively magnetic resonance imaging (MRI)-defined bone marrow lesions (BMLs) and inflammatory markers is associated with change in quantitatively-assessed cartilage loss in the femorotibial joint (FTJ) in knees with radiographic osteoarthritis (OA) over 24 months.DesignParticipants were included from the IMI-APPROACH and the Osteoarthritis Initiative FNIH studies. Semiquantitative MRI assessment was performed … Continued

Comparison between coronal FLASH and sagittal double echo steady state MRI in detecting longitudinal cartilage thickness change by fully automated segmentation – data from the FNIH biomarker cohort

Objective: Artificial intelligence (AI-) based automated cartilage analysis demonstrated similar sensitivity to change and only slighty inferior differentiation between radiographic progressors and non-progressors compared with manual segmentation. However, this finding was based on DESS MRI from the Osteoarthritis Initiative (OAI), whereas the vast majority of multicenter clinical trials rely on T1-weighted gradient echo (e.g. FLASH). Here … Continued

Cartilage loss is greater in knees with virtual joint replacement status than in matched controls without

Objective: We examined whether the trajectory of femorotibial cartilage loss differs between knees meeting a clinically defined virtual knee replacement (vKR) status based on patient-reported outcomes vs. those with low probability. Design: vKR cases (highest 10 ​% of probabilities for having surgical KR) were selected using knee pain and quality of life criteria, developed from knees with … Continued

A fully-automated technique for cartilage morphometry in knees with severe radiographic osteoarthritis – method development and validation

Objective: Denuded areas of subchondral bone (dAB) pose a challenge for fully automated segmentation of articular cartilage and subchondral bone in knees with severe radiographic osteoarthritis using convolutional neural networks (CNNs). Here we propose an automated post-processing relying on a selection-based multi-atlas registration for reconstructing the total area of subchondral bone (tAB) to overcome this issue. … Continued

Impact of quantified knee positioning on the measurement of minimal joint space width using statistical shape models: A cross-sectional and longitudinal analysis in the IMI-APPROACH.

Objective: Correlations between minimum joint space width (mJSW) and MRI-based cartilage thickness are strong in cross-sectional analyses and moderate in longitudinal analyses, possibly due to knee rotation and flexion. This study investigates the effect of knee positioning during radiographic acquisition on the difference between mJSW and MRI-based cartilage thickness. Methods: Radiographic mJSW from the index … Continued

Clinical validation of fully automated cartilage transverse relaxation time (T2) and thickness analysis using quantitative DESS magnetic resonance imaging

Objective: To clinically validate a fully automated cartilage segmentation technique from quantitative double-echo steady-state (qDESS) MRI supporting simultaneous estimation of cartilage T2 and morphology. Here, we test whether laminar (superficial and deep layer) T2 results from convolutional neural network (CNN) segmentations are consistent with those from manual expert segmentations. Materials and methods: The 3D qDESS sequence was … Continued