Osteoarthritis imaging最新文献

{"title":"ACETABULAR DYSPLASIA TRAJECTORIES DURING GROWTH FROM 6 TO 18 YEARS","authors":"N. Hendriks,&nbsp;Z. Zheng,&nbsp;F. Boel,&nbsp;F. Rivadeneira,&nbsp;C.J. Tiderius,&nbsp;S.M.A. Bierma-Zeinstra,&nbsp;R. Agricola,&nbsp;J. Runhaar","doi":"10.1016/j.ostima.2026.100428","DOIUrl":"10.1016/j.ostima.2026.100428","url":null,"abstract":"<div><h3>INTRODUCTION</h3><div>Acetabular dysplasia (AD) is an important risk factor for early onset hip OA in relatively young individuals. AD is defined using various hip morphology measures. However, the developmental course of these measures from childhood to adulthood remains unclear. Understanding their longitudinal patterns is therefore essential.</div></div><div><h3>OBJECTIVE</h3><div>To describe the longitudinal course of certain validated AD measures and to identify distinct trajectories of these measures.</div></div><div><h3>METHODS</h3><div>Data from the population-based Generation R Study were used. Participants were examined around age 6, 9, 13 and 18 years. Participants with high-resolution anteroposterior (AP) dual-energy x-ray absorptiometry (DXA) of the right hip and AP full-body DXA were included. In total, 6332 participants were included (3107 males and 3225 females). Hip shape was outlined with radiological landmarks used to calculate the lateral center edge angle (LCEA; indicator of acetabular coverage), and the acetabular depth-width ratio (ADR; quantifying acetabular depth). The course of LCEA and ADR was assessed using all available data. Latent class growth analysis (LCGA) was used to identify trajectories, including participants with at least two measurements.</div></div><div><h3>RESULTS</h3><div>Both LCEA and ADR increased from ages 6 to 18, reaching mean values of 29.9° (SD 5.8) for LCEA and 326.4 (SD 31.8) for ADR. In both measures the largest increase was between 9 and 13 years. Table 1 shows all the mean values and SDs of each AD measure for all time points for males and females separately. In the LCGA, 3868 participants were included. No significant trajectories were identified for LCEA. For ADR, a quadratic two-class model best fit the data in both sexes. In males, mean ADR at age 18 was 380.4 (SD 23.3) in the trajectory with highest increase (Class 1, n = 199), and 317.0 (SD 26.0) in the other trajectory (Class 2, n = 1728). In females, corresponding values were 353.4 (SD 20.5) in the trajectory with highest increase (Class 1, n = 808), and 310.3 (SD 22.0) in the other trajectory (Class 2, n = 1321) (Figure 1).</div></div><div><h3>CONCLUSION</h3><div>This study describes the longitudinal course of the LCEA and ADR from childhood to early adulthood. While no distinct LCEA trajectories were observed, two ADR trajectories were identified in both sexes, mainly differing in acetabular depth at age 18. The smaller classes showed a deeper acetabulum during growth. Lower ADR values, indicating a shallower acetabulum, and are associated with AD. Understanding of acetabular development may support earlier identification of individuals at risk for AD-related hip OA.</div></div>","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100428"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"COMBINING MOAKS AND MRI RADIOMICS IMPROVES MACHINE-LEARNING PREDICTION OF RADIOGRAPHIC KNEE OA PROGRESSION: DATA FROM THE FNIH","authors":"N. Eshraghi,&nbsp;P. Mirghaderi,&nbsp;Y. Xu,&nbsp;P.C. Thurlow,&nbsp;M. Nyflot,&nbsp;J.S. Wu,&nbsp;T.M. Link,&nbsp;A. Guermazi,&nbsp;M. Chalian","doi":"10.1016/j.ostima.2026.100425","DOIUrl":"10.1016/j.ostima.2026.100425","url":null,"abstract":"<div><h3>INTRODUCTION</h3><div>Radiographic knee OA progression is heterogeneous and may not be well predicted by clinical and biospecimen biomarkers alone. Baseline MRI-derived features, including expert semiquantitative MOAKS and quantitative radiomics, may capture complementary whole-joint structural information relevant to progression risk.</div></div><div><h3>OBJECTIVE</h3><div>To develop and compare multimodal machine-learning models for predicting 24-month radiographic knee OA progression using clinical variables, biospecimen biomarkers, semiquantitative MRI features, and MRI radiomics from the FNIH Osteoarthritis Biomarkers Consortium.</div></div><div><h3>METHODS</h3><div>This nested case-control analysis included 600 OAI participants (600 knees) with baseline KL grade 1-3, baseline knee MRI, biospecimen biomarkers, and baseline and 24-month radiographs. Radiographic progression was defined as a decrease of at least 0.7 mm in minimum medial tibiofemoral JSW from baseline to 24 months. Baseline predictors included 12 clinical variables, 18 biospecimen biomarkers (Serum/urine), 78 MOAKS features, and 400 radiomics features extracted from femoral/tibial cartilage and bone on sagittal 3D DESS MRI after automated segmentation. Four feature sets were tested: BioClinical (M1), MOAKS (M2), Radiomics (M3), and Combined (M4). Logistic regression, XGBoost, random forest, and SVM were evaluated using nested cross-validation with pooled out-of-fold AUC, sensitivity, and specificity. AUC differences were tested using bootstrap resampling.</div></div><div><h3>RESULTS</h3><div>Among 600 participants, 297 (49.5%) developed radiographic progression; the BioClinical model showed modest discrimination (AUC 0.57-0.59). Adding MOAKS features improved AUC to 0.71-0.74, and adding radiomics improved AUC to 0.69-0.71; both outperformed BioClinical models across all algorithms (all p&lt;0.0001). MOAKS and Radiomics models did not significantly differ from each other (all p&gt;0.05). The Combined model achieved the best overall performance (AUC 0.75-0.77), with XGBoost and SVM reaching AUC 0.77. Combined models improved sensitivity/specificity for logistic regression from 0.64/0.50 to 0.70/0.71 and for SVM from 0.49/0.57 to 0.70/0.73. Decision curve analysis showed that Combined models provided the highest or near-highest net benefit across most threshold probabilities.</div></div><div><h3>CONCLUSION</h3><div>Baseline MRI-derived features substantially improved machine-learning prediction of radiographic knee OA progression beyond clinical and biospecimen biomarkers alone. MOAKS and radiomics provided broadly comparable predictive performance, while their integration achieved the best overall discrimination, supporting complementary roles for expert semiquantitative scoring and automated quantitative MRI radiomics in OA progression risk stratification.</div></div>","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100425"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ARTERIAL REMODELING ON KNEE MRI AND CARTILAGE T2 RELAXATION TIME: DATA FROM THE OSTEOARTHRITIS INITIATIVE","authors":"N. Eshraghi,&nbsp;P. Mirghaderi,&nbsp;F. Pishgar,&nbsp;N. Balu,&nbsp;C. Yuan,&nbsp;A. Guermazi,&nbsp;M. Chalian","doi":"10.1016/j.ostima.2026.100424","DOIUrl":"10.1016/j.ostima.2026.100424","url":null,"abstract":"<div><h3>INTRODUCTION</h3><div>Popliteal artery remodeling may contribute to knee OA through altered periarticular and subchondral perfusion. Cartilage T2 relaxation time is a quantitative MRI biomarker of cartilage composition, reflecting collagen organization and water content. Whether automated MRI-derived popliteal artery parameters are associated with cartilage T2 relaxation time and short-term T2 change remains unclear.</div></div><div><h3>OBJECTIVE</h3><div>To evaluate cross-sectional and longitudinal associations between baseline MRI-derived popliteal artery parameters and cartilage T2 relaxation time in OAI participants.</div></div><div><h3>METHODS</h3><div>This retrospective longitudinal analysis included 4,447 OAI participants with baseline right-knee cartilage T2 measurements; 3,854 and 3,560 participants had 12- and 24-month follow-up T2 data, respectively. Popliteal artery lumen and wall contours were derived from 3T sagittal DESS MRI using a fully automated segmentation pipeline. Arterial parameters included mean wall thickness, mean wall area, mean lumen area, mean total artery area, eccentricity ratio (maximum-to-minimum wall thickness ratio), and normalized wall index (NWI; wall area divided by total vessel area). Cartilage T2 values were obtained from automated segmentation of multi-echo spin-echo MRI in the lateral tibia, lateral femur, medial tibia, medial femur, and patella. Adjusted linear regression models tested associations between standardized arterial parameters and baseline T2 and 12-/24-month T2 change. Models adjusted for age, sex, BMI, smoking, diabetes, PASE, and comorbidity score; longitudinal models additionally adjusted for baseline T2. Analyses were stratified by KL grade &lt;2 versus ≥2. Benjamini-Hochberg FDR-adjusted p&lt;0.05 was considered significant.</div></div><div><h3>RESULTS</h3><div>Mean age was 61.1 ± 9.1 years, 2,612/4,447 (58.7%) participants were female, and mean BMI was 28.5 ± 4.8 kg/m2. Mean wall area, mean lumen area, and mean total artery area were consistently associated with higher baseline cartilage T2 values across all subregions (β=0.25-0.45 per SD; all p&lt;0.001). Mean NWI was inversely associated with baseline T2 across all subregions (β=-0.15 to -0.33 per SD; all p&lt;0.001), with the strongest association in the patella (β=-0.33, 95% CI=-0.45 to -0.22; p&lt;0.001). Mean wall thickness was associated only with lateral tibial (β=0.13, 95% CI=0.04 to 0.23; p=0.006) and lateral femoral T2 (β=0.11, 95% CI=0.01 to 0.21; p=0.039). Baseline arterial parameters were not associated with 12- or 24-month T2 change after baseline T2 adjustment. Effect sizes were generally larger in KL≥2.</div></div><div><h3>CONCLUSION</h3><div>Automated knee MRI-derived popliteal artery size and remodeling parameters were associated with baseline cartilage T2 relaxation time, particularly in knees with radiographic OA, but did not predict short-term T2 change.</div></div>","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100424"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LONGITUDINAL JOINT SPACE WIDTH MEASURES AND RISK OF KNEE REPLACEMENT IN PATIENTS WITH SYMPTOMATIC AND RADIOGRAPHIC KNEE OSTEOARTHRITIS: DATA FROM OSTEOARTHRITIS INITIATIVE","authors":"H Cen,&nbsp;Q Dang,&nbsp;A Akba,&nbsp;Z Wang,&nbsp;H Chen,&nbsp;Y Zhang,&nbsp;G Ruan,&nbsp;P Cao,&nbsp;Z Zhu,&nbsp;C Ding","doi":"10.1016/j.ostima.2026.100427","DOIUrl":"10.1016/j.ostima.2026.100427","url":null,"abstract":"<div><h3>INTRODUCTION</h3><div>Although previous studies have shown that baseline joint space width (JSW) and its short-term change are associated with knee replacement (KR) in patients with knee osteoarthritis (OA), it remains unclear whether the longitudinal trajectory of JSW is associated with KR risk.</div></div><div><h3>OBJECTIVE</h3><div>To evaluate the associations of longitudinal trajectories of minimum JSW (mJSW) and JSW at x = 0.250 (JSW250) in the medial tibiofemoral compartment with the risk of KR among patients with symptomatic (frequent knee symptoms for at least one month in the past 12 months) and radiographic knee OA (Kellgren–Lawrence grade 2 or 3).</div></div><div><h3>METHODS</h3><div>Using data from the Osteoarthritis Initiative (OAI) database, in which JSW was measured from baseline to the 4th visit, and again at the 6th and 8th visits, with KR status recorded over 9 visits, we applied a Bayesian joint model of longitudinal and time-to-event data to determine the association of longitudinal mJSW and JSW250 with KR risk. JSW measures were standardized to obtain hazard ratios (HRs) per 1‑SD increase. Joint models are preferred for simultaneous analysis of repeated measurements and survival data, providing less biased estimates.</div></div><div><h3>RESULTS</h3><div>A total of 729 participants with baseline and at least one follow‑up measurement for both JSW metrics were included. After a median follow‑up of 8.84 years, 118 participants (15.97%) underwent KR. The Bayesian joint model showed that, at any time point, higher values of both measures were associated with a reduced risk of KR. Specifically, the HR per 1‑SD increase was 0.70 (95% credible interval [CrI]: 0.56–0.86) for mJSW and 0.60 (95% CrI: 0.49–0.75) for JSW250, indicating a stronger protective association for JSW250.</div></div><div><h3>CONCLUSION</h3><div>Repeated measurements of mJSW and JSW250 provide strong longitudinal associations with KR risk, with JSW250 showing a more pronounced effect. These findings underscore the value of regular monitoring of JSW measures, particularly JSW250, in clinical practice to facilitate timely identification of knee OA patients at high risk for KR.</div></div>","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100427"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543768","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Posters","authors":"","doi":"10.1016/S2772-6541(26)00071-1","DOIUrl":"10.1016/S2772-6541(26)00071-1","url":null,"abstract":"","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100461"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AUTOMATIC FEMOROTIBIAL JOINT SPACE PATCH SELECTION FROM WEIGHT-BEARING CONE-BEAM COMPUTED TOMOGRAPHY BONE SEGMENTATIONS","authors":"T.J. Numminen,&nbsp;L. Vuononvirta,&nbsp;T. Frondelius,&nbsp;T. Turmezei,&nbsp;M.T. Nevalainen,&nbsp;J. Määttä,&nbsp;S. Meriläinen,&nbsp;P. Lehenkari,&nbsp;M.A.J. Finnilä,&nbsp;S.J.O. Rytky","doi":"10.1016/j.ostima.2026.100438","DOIUrl":"10.1016/j.ostima.2026.100438","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;INTRODUCTION&lt;/h3&gt;&lt;div&gt;Weight-bearing cone-beam CT (CBCT) enables evaluation of the knee joint under physiological load in a standing position. It has been used to assess OA-induced structural changes in the joint with a higher sensitivity compared to conventional radiography. However, labor-intensive manual bone and femorotibial joint space patch (FT-JSP) segmentation is currently required to quantify the 3D anatomy and joint space width (JSW) of the load-bearing region with CBCT. Here, we implemented a deep learning model for automatic and accurate FT-JSP selection.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;OBJECTIVE&lt;/h3&gt;&lt;div&gt;To develop a deep learning model to segment FT-JSP from 2D map representations derived from bone segmentations.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;METHODS&lt;/h3&gt;&lt;div&gt;The dataset comprised weight-bearing CBCT scans from a weight-loss surgery study acquired at multiple time points. Imaging was performed using the Planmed Verity system with 0.2 mm isotropic voxel size, 96 kVp tube voltage, 12 mA current, and 6 s acquisition time. Of 264 scans from 86 patients, four were excluded due to errors in the generated femur/tibia objects, leaving 260 scans for analysis. The femur, tibia, and FT-JSP were manually segmented using Stradview, and the FT-JSP segmentation was used as a gold standard. From these segmentations, 2D geometric representations were derived, including a heightmap calculated 10 mm above the highest position of the femur and an FT distance map. Data were split at the patient level into training (n=176), validation (n=31), and hold-out test (n=53) sets to prevent leakage across repeated time points. A residual 2D U-Net implemented in the PyTorch-based MONAI framework was trained to segment the FT-JSP from these derived 2D maps, and the prediction was thresholded at 0.8 to obtain a binary segmentation mask. The performance of the method was evaluated using the Dice score, Jaccard index, precision, and recall. Finally, the segmentation was projected on the femur 3D surface to evaluate the 95&lt;sup&gt;th&lt;/sup&gt; percentile Hausdorff distance (HD95).&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;RESULTS&lt;/h3&gt;&lt;div&gt;The automatic FT-JSP segmentation achieved a Dice score of 0.90 ± 0.03, 0.89 ± 0.05, 0.88 ± 0.03, and a Jaccard index of 0.83 ± 0.04, 0.80 ± 0.07, 0.79 ± 0.05 for training, validation, and hold-out sets, respectively (Table 1). On the hold-out test set, precision and recall were 0.89 and 0.89, respectively. After reconstructing the automatically obtained FT-JSP masks into a 3D surface representation, the HD95 to the manual segmentation was 1.96 ± 0.21 mm, with a representative qualitative example shown in Figure 1.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;CONCLUSION&lt;/h3&gt;&lt;div&gt;A residual 2D U-Net can automatically segment the FT-JSP from weight-bearing CBCT-derived geometric representations. These results support the feasibility of the method for automated FT-JSP identification and show that the predicted 2D regions can be reconstructed into 3D space with good anatomical agreement. The next steps i","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100438"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CYTOKINE DYNAMICS AND MRI-BASED CARTILAGE CHANGES OVER 5 YEARS AFTER ACL INJURY","authors":"S. Larsson,&nbsp;F. Eckstein,&nbsp;L.S. Lohmander,&nbsp;A. Struglics","doi":"10.1016/j.ostima.2026.100410","DOIUrl":"10.1016/j.ostima.2026.100410","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;INTRODUCTION&lt;/h3&gt;&lt;div&gt;Anterior cruciate ligament (ACL) injury triggers an acute intra-articular inflammatory response, characterized by elevated synovial fluid levels of pro‑ and anti‑inflammatory cytokines that decline over time. These cytokines collectively influence the balance between cartilage breakdown and repair.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;OBJECTIVE&lt;/h3&gt;&lt;div&gt;In this exploratory study, we examined whether cytokine concentrations at different time points after ACL injury were associated with subsequent changes in cartilage thickness, and identified time periods with the strongest associations.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;METHODS&lt;/h3&gt;&lt;div&gt;The KANON cohort includes patients with an acute ACL tear in a previously uninjured knee (n=121; mean age 26 years at injury; 74% men). Synovial fluids were collected at baseline (0 to 6 weeks after injury) and at follow-ups at 16, 30, 52, 104, and 260 weeks. Cytokines IL-6, IL-8, IL-10, and TNF were quantified using multiplex immunoassays, and IL-1α and LAP TGFβ-1 by proximity extension assay by Olink. MRI-based cartilage thickness was measured in 16 femorotibial cartilage subregions within each knee at baseline, 2 and 5 years using quality-controlled manual segmentation of 3D SPGR images (1.5 T Philips Gyroscan). A location-independent subregional cartilage thickness change score (ChangeScore) was calculated as the sum of all absolute changes without taking the sign (±) into account. Associations between cytokine concentrations at each visit and ChangeScores over two periods (baseline to 2 years, and 2 to 5 years) were examined using linear regression models adjusted for age and BMI at baseline, and sex. Variables were standardized (z-scores) to allow comparison of effect sizes. No adjustments for multiplicity were made.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;RESULTS&lt;/h3&gt;&lt;div&gt;&lt;strong&gt;Early (0-2 years)&lt;/strong&gt; Despite markedly elevated cytokine concentrations in the acute post-injury phase &lt;strong&gt;(Figure 1A)&lt;/strong&gt;, baseline cytokine levels were not significantly associated with cartilage ChangeScore 0 to 2 years &lt;strong&gt;(Figure 1B).&lt;/strong&gt; However, cytokine concentrations at 16, 30, and 52 weeks showed partial statistically significant associations with cartilage ChangeScore during the first 2 years, suggesting that subacute inflammation may be related to early cartilage matrix loss and gain.&lt;/div&gt;&lt;div&gt;&lt;strong&gt;Late (2-5 years)&lt;/strong&gt; Cytokine levels measured during the first year were not associated with cartilage changes between 2 and 5 years, whereas cytokine concentrations at the 2- and 5-year visits showed modest but statistically significant associations with ChangeScore 2 to 5 years, indicating a link between persistent low-grade inflammation and ongoing change. Among the six cytokines, IL-6 and IL-10 showed the largest effect sizes per standard deviation change in concentration and the most persistent associations with cartilage thickness ChangeScores across both time periods, whereas IL-1α displayed the weakest and","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100410"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PERIPHERAL-TO-CENTRAL TRABECULAR BONE TEXTURE IS ASSOCIATED WITH STRUCTURAL PROGRESSION OF KNEE OSTEOARTHRITIS: LONGITUDINAL DATA FROM THE MOST COHORT","authors":"P. Podsiadlo,&nbsp;M. Wolski,&nbsp;J. Crawmer,&nbsp;R. Tang,&nbsp;C.K. Kwoh,&nbsp;T. Neogi,&nbsp;J. Duryea","doi":"10.1016/j.ostima.2026.100451","DOIUrl":"10.1016/j.ostima.2026.100451","url":null,"abstract":"<div><h3>INTRODUCTION</h3><div>Bone morphology is increasingly recognized as important in the pathogenesis and progression of OA. Previous studies have shown that medial subchondral trabecular bone texture (TBT) predicts OA progression. However, the predictive validity of peripheral-to-central medial TBT regions as a structural imaging biomarker for radiographic knee OA progression is not clear.</div></div><div><h3>OBJECTIVE</h3><div>To investigate the predictive validity of longitudinal changes in TBT across peripheral-to-central medial subchondral regions assessed using directional fractal dimensions</div></div><div><h3>METHODS</h3><div>We used data from the MOST (Multicenter Osteoarthritis Study), right knees from the Birmingham, Alabama clinical site at baseline and 60-month follow-up (n = 726). Regions of interest (ROIs) were defined with anatomical landmarks to ensure consistent size and position (Figure 1). Three regions were placed immediately beneath the medial tibial cortical plate, extending from the peripheral medial edge (ROI 1) towards the intermediate (ROI 2) and central medial tibia (ROI 3). For each region, fractal dimensions (FD) along the horizontal (FD_H), vertical (FD_V) and roughest (FD_Sta) directions were calculated. Higher FD values indicate rougher bone texture (i.e., more complex and irregular bone architecture), and the highest value defines FD along the roughest direction. TBT assessments were performed at baseline and 60 months. Knees that remained KL 0/1 at 60-months served as the control group. Comparisons were performed between the control group (n = 474) and knees that progressed from KL 2 to 3 (n = 50), KL 2 to 4 (n = 9), or KL 3 to 4 (n = 56). Differences in mean 60-month FD values between the progression and control groups were assessed using linear regression, adjusted for baseline FD values. Comparisons were based on estimated marginal means with Dunnett adjustment for multiple comparisons.</div></div><div><h3>RESULTS</h3><div>Results are summarized in Table 1. Vertical FD showed the strongest and most consistent differences between progressor and control knees across ROIs. FD_Sta and FD_V were significantly lower in knees progressing from KL 3 to 4 across all regions, and from KL 2 to 4 in ROI 1 and ROI 2. For KL 2 to 3 progression, FD_V in ROI 2 and ROI 3 and FD_Sta in ROI 3 differed significantly, whereas FD_H was significant only for KL 3 to 4 progression.</div></div><div><h3>CONCLUSION</h3><div>The risk of radiographic progression to advanced OA was associated with lower medial trabecular bone texture complexity, with the vertical direction showing the strongest and most consistent differences across all regions. These findings suggest that region-specific directional fractal bone texture analysis may serve as a potential structural imaging biomarker for identifying knees at risk of radiographic OA progression.</div></div>","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100451"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"REGIONAL ANALYSIS OF LATERAL MENISCUS MECHANICS IN CADAVERIC KNEES USING A NOVEL ULTRA-HIGH FIELD MRI METHOD","authors":"J.S. Broberg,&nbsp;E. Hoptioncann,&nbsp;A. Kimbowa,&nbsp;A. Yung,&nbsp;K. Bale,&nbsp;I. Hacihaliloglu,&nbsp;P. Lodhia,&nbsp;D.R. Wilson","doi":"10.1016/j.ostima.2026.100416","DOIUrl":"10.1016/j.ostima.2026.100416","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;INTRODUCTION&lt;/h3&gt;&lt;div&gt;The lateral meniscus is an important load distributor in the knee. Injuries or degradation of the meniscus can disrupt knee mechanics, which can lead to increased stress on the cartilage and eventual development of OA. Clinically, meniscal extrusion can be measured with MRI and is often used as a surrogate for disrupted meniscal mechanics. It is used to help diagnose meniscal pathology; however, these assessments are often done only in the coronal plane with the knee in a supine and unloaded posture. Furthermore, clinical 3T MRI scanners lack the image resolution to accurately measure strains in the meniscus and cartilage. Thus, the link between extrusion and loaded knee mechanics is unclear. As a result, surgeons are uncertain how important it is to restore extrusion when repairing the meniscus with procedures like centralization. Ultra-high field MRI provides sufficient resolution to measure both cartilage and meniscal mechanics during loading, without disruption or simulation of the articulating joint surfaces.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;OBJECTIVE&lt;/h3&gt;&lt;div&gt;Determine where in the meniscus extrusion changes the most under load, and how this relates to meniscal strain. We hypothesize that meniscal extrusion will increase the most in the body region, and that the region with the greatest extrusion will have the largest strains.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;METHODS&lt;/h3&gt;&lt;div&gt;Six human knee lateral compartments (mean age 70 yrs) with ACL and meniscus preserved were tested. Specimens were placed in a novel pneumatic compression apparatus customized for use in a 9.4T MRI scanner. Morphologic FLASH and RARE scans (axial, sagittal, and coronal resolution of 0.06 × 0.12 × 0.4 mm) were acquired before loading and after 2 hours of 48% body weight loading. The change in meniscal extrusion from unloaded to loaded states was calculated in the radial direction, at every degree around the meniscus circumference. For the strain analysis, the MR images were resampled to 0.2 mm isotropic resolution. The Advanced Normalization Tools module in 3D Slicer was used to register the resampled loaded image to the resampled unloaded image. Gradients of the resulting displacement field from the image registration were then used to calculate meniscal strains in the radial, circumferential, and axial directions. Average strains were determined for six different regions of the meniscus: internal and external anterior horn, internal and external body, and internal and external posterior horn. Repeated one-way ANOVA tests with Tukey’s multiple comparisons were used to compare the strains in different regions, and to compare extrusions at different 15° increments around the circumference of the meniscus.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;RESULTS&lt;/h3&gt;&lt;div&gt;Loading produced the greatest extrusion changes in the anterior horn and the smallest extrusion changes in the meniscus body (Figure 1). Change in meniscal extrusion after loading was greater at 60° than at 15° (p = 0.011) or 30°","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100416"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"nnU-NET BASED 3D SEGMENTATION OF HIGH-RESOLUTION KNEE CT SCANS FOR THE ANALYSIS OF SHAPE AND INTRA-ARTICULAR CALCIFICIATIONS.","authors":"H. Park,&nbsp;W. Wang,&nbsp;J.A. Lynch,&nbsp;T.M. Link,&nbsp;T. Neogi","doi":"10.1016/j.ostima.2026.100413","DOIUrl":"10.1016/j.ostima.2026.100413","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;INTRODUCTION&lt;/h3&gt;&lt;div&gt;High-resolution knee CT scans provide fine detail about the 3D shape of the bones in the knee, but manually segmenting multiple slices in fine detail can be very tedious. Intra-articular mineralization quantification could potentially be a useful imaging biomarker, but its quantification would require masking out the signal from bone mineral. Therefore, an automated method of accurately defining the extent of the femur, tibia, and patella on knee CT potentially has multiple uses.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;OBJECTIVE&lt;/h3&gt;&lt;div&gt;To develop an automated tool to segment the femur, tibia, and patella from knee CT scans.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;METHODS&lt;/h3&gt;&lt;div&gt;From 4,014 knee CT scans acquired in the MOST Study at Year 12, 234 knees were selected to represent a range of osteoarthritis (OA) severity based on tibiofemoral (TF) and patellofemoral (PF) Kellgren–Lawrence grades (KLG). Knees were categorized as”&lt;/div&gt;&lt;div&gt; &lt;!--&gt;(1) no OA (TF KLG &lt; 2, PF KLG &lt; 2)&lt;/div&gt;&lt;div&gt; &lt;!--&gt;(2) TF OA only (TF KLG ≥ 2, PF KLG &lt; 2)&lt;/div&gt;&lt;div&gt; &lt;!--&gt;(3) PF OA only (TF KLG &lt; 2, PF KLG ≥ 2)&lt;/div&gt;&lt;div&gt; &lt;!--&gt;(4) mixed OA (TF KLG ≥ 2, PF KLG ≥2 2)&lt;/div&gt;&lt;div&gt;TF OA cases were further classified as medial or lateral using OARSI joint space narrowing grades. The sample was balanced across OA categories, TF compartment involvement, and KLG distributions. Femur, tibia, and patella were manually segmented and used to train and validate a deep learning model based on the nnU-Net framework. Images were acquired with a standardized protocol and left unnormalized, but trimmed to a standardized 450 axial slices centered around the femoral epicondyles to give 3D image volumes of size 512 × 512 × 450.&lt;/div&gt;&lt;div&gt;Images were split into training and validation sets for 5-fold cross-validation, but stratified to ensure equal numbers of knees from each OA class in each of the 5 splits. We also made sure that if a study participant contributed 2 knees to the sample, both knees would be in the same split. For the 5-fold cross-validation, we used a 3d nnU-Net v2 cascaded architecture combining coarse localization trained on an entire low-resolution 3D volume with a subsequent patch-based full-resolution refinement.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;RESULTS&lt;/h3&gt;&lt;div&gt;Initial runs of the low-resolution 3D training showed no signs of overfitting or underfitting by 500 epochs, so the entire cascade was performed using each of the 5 training sets, and then the 5 models were used to predict segmentations for their matching validation sets. Table 1 shows the results for agreement between the manual segmentations and the predicted segmentations presented as DICE coefficients. Figure 1 shows an example of a predicted segmentation.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;CONCLUSION&lt;/h3&gt;&lt;div&gt;The nnU-NET framework provides a simple and reliable method for developing a deep learning model for segmenting femur, tibia and patella from CT scans of the knee. This model was developed on high resolution unilateral reconstru","PeriodicalId":74378,"journal":{"name":"Osteoarthritis imaging","volume":"6 ","pages":"Article 100413"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148543814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}