Dataset Information

A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction.

ABSTRACT: Grading individual knee osteoarthritis (OA) features is a fine-grained knee OA severity assessment. Existing methods ignore following problems: (1) more accurately located knee joints benefit subsequent grades prediction; (2) they do not consider knee joints' symmetry and semantic information, which help to improve grades prediction performance. To this end, we propose a SE-ResNext50-32x4d-based Siamese network with adaptive gated feature fusion method to simultaneously assess eight tasks. In our method, two cascaded small convolution neural networks are designed to locate more accurate knee joints. Detected knee joints are further cropped and split into left and right patches via their symmetry, which are fed into SE-ResNext50-32x4d-based Siamese network with shared weights, extracting more detailed knee features. The adaptive gated feature fusion method is used to capture richer semantic information for better feature representation here. Meanwhile, knee OA/non-knee OA classification task is added, helping extract richer features. We specially introduce a new evaluation metric (top±1 accuracy) aiming to measure model performance with ambiguous data labels. Our model is evaluated on two public datasets: OAI and MOST datasets, achieving the state-of-the-art results comparing to competing approaches. It has the potential to be a tool to assist clinical decision making.

SUBMITTER: Wang K

PROVIDER: S-EPMC8376929 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundLimiting treatment to those recommended by the American Academy of Orthopaedic Surgeon Clinical Practice Guidelines has been suggested to decrease costs by 45% in the year prior to total knee arthroplasty, but this only focuses on expenditures leading up to, but not including, the surgery and not the entire episode of care. We evaluated the treatment costs following knee osteoarthritis (OA) diagnosis and determined whether these are different for patients who use intra-articular hyaluronic acid (HA) and/or knee arthroplasty.MethodsClaims data from a large commercial database containing de-identified data of more than 100 million patients with continuous coverage from 2012 to 2016 was used to evaluate the cumulative cost of care for over 2 million de-identified members with knee OA over a 4.5-year period between 2011 and 2015. Median cumulative costs were then stratified for patients with or without HA and/or knee arthroplasty.ResultsKnee OA treatment costs for 1,567,024 patients over the 4.5-year period was $6.60 billion (mean $4210/patient) as calculated by the authors. HA and knee arthroplasty accounted for 3.0 and 61.5% of the overall costs, respectively. For patients who underwent knee arthroplasty, a spike in median costs occurred sooner for patients without HA use (around the 5- to 6-month time point) compared to patients treated with HA (around the 16- to 17-month time point).ConclusionsNon-arthroplasty therapies, as calculated by the authors, accounted for about one third of the costs in treating knee OA in our cohort. Although some have theorized that limiting the use of HA may reduce the costs of OA treatment, HA only comprised a small fraction (3%) of the overall costs. Among patients who underwent knee arthroplasty, those treated with HA experienced elevated costs from the surgery later than those without HA, which reflects their longer time to undergoing knee arthroplasty. The ability to delay or avoid knee arthroplasty altogether can have a substantial impact on the cost to the healthcare system.

Project description:IntroductionObservational data suggest that vitamin D deficiency is associated with the onset and progression of knee osteoarthritis (OA). However, randomised controlled trials (RCTs) to date investigating the efficacy of vitamin D supplementation in knee OA have reported conflicting results. Further research is needed to clarify the effects of vitamin D on patient-reported outcomes and determine whether there are patient subgroups who may benefit from the supplementation. The aim of this individual patient data (IPD) meta-analysis is to identify patient-level predictors of treatment response to vitamin D supplementation on pain and physical function.Methods and analysisA systematic literature search will be conducted for RCTs of vitamin D supplementation on knee OA. Authors of original RCTs will be contacted to obtain the IPD. The primary outcomes will include long-term (≥12 months) pain and physical function. Secondary outcomes will include medium-term (≥6 months and <12 months) and short-term (<6 months) pain and physical function, as well as patient global assessment, quality of life and adverse events. Potential treatment effect modifiers to be examined in the subgroup analyses include age, gender, body mass index, baseline knee pain severity and physical function, baseline vitamin D level, radiographic stage, presence of bone marrow lesions on MRI, presence of clinical signs of local inflammation and concomitant depressive symptoms. Both one-step and two-step modelling methods will be used to determine the possible modifiable effect of each subgroup of interest.Ethics and disseminationResearch ethical or governance approval is exempt for this study as no new data are being collected. This study will be the first IPD meta-analysis to clarify the effect of vitamin D supplementation on clinical symptoms in different subgroups of patients with knee OA. The findings will be disseminated through peer-review publications and conference presentations.Prospero registration numberCRD42018107740.

Project description:BackgroundSynovitis occurring frequently in osteoarthritis (OA) may be a targeted outcome. There are no data examining whether synovitis changes following intra-articular intervention.MethodsPersons aged 40 years and older with painful knee OA participated in an open label trial of intra-articular steroid therapy. At all time points they completed the Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire. They had a contrast-enhanced (CE) MRI immediately prior to an intra-articular steroid injection with a repeat scan within 20 days. Response status was assessed using the Osteoarthritis Research Society International (OARSI) response criteria. OARSI responders were followed until their pain relapsed either within 20% of baseline or 6 months, shortly after which a third MRI was performed. Synovial tissue volume (STV) was measured on postcontrast knee images. We looked at changes in the STV and in pain, and their association.Results120 subjects with preinjection and postinjection CE MRI were followed. Their mean age was 62.3 years (SD=10.3) and 62 (52%) were women. The median time between injection and follow-up scan was 8 days (IQR 7-14 days). 85/120 (71%) were OARSI responders. Pain decreased (mean change in KOOS=+23.9; 95% CI 20.1 to 27.8, p<0.001) following steroid injection, as did mean STV (mean change=-1071 mm(3); 95% CI -1839 mm(3) to -303 mm(3), p=0.01). Of the 80 who returned for a third MRI, pain relapsed in 57, and in the 48 of those with MRI data, STV increased between follow-up and final visit (+1220 mm(3); 95% CI 25 mm(3) to 2414 mm(3), p=0.05). 23 were persistent responders at 6 months and, in these, STV did not increase (mean change=-202 mm(3); 95% CI -2008 mm(3) to 1604 mm(3), p=0.83). Controlling for variation over time, there was a significant association between synovitis volume and KOOS pain (b coefficient-change in KOOS pain score per 1000 mm(3) change in STV=-1.13; 95% CI -1.87 to -0.39, p=0.003), although STV accounted for only a small proportion of the variance in change in pain.ConclusionsSynovial tissue volume in knee OA shrinks following steroid therapy, and rebounds in those whose pain relapses. It can be considered a treatment target in symptomatic knee OA.Trial registration numberISRCTN07329370.

Project description:INTRODUCTION:Randomised clinical trials to date investigating the efficacy of bisphosphonates in knee osteoarthritis (OA) have found divergent results, with a recent meta-analysis finding no superiority of these drugs over placebo. Whether particular patient subgroups are more likely to benefit from this therapy than others is still unclear. We aim to investigate the effects of bisphosphonates compared with a control group (placebo, no treatment, another active treatment) on clinical and structural outcomes in specific knee OA subpopulations with possible distinct rates of subchondral bone turnover. METHODS AND ANALYSIS:Medline, Embase, Scopus, Web of Sciences and Cochrane Central Register of Controlled Trials will be searched from inception to February 2018. Randomised clinical trials will be eligible if they reported at least one potential treatment effect modifier at baseline: gender, menopausal status, age, body mass index, radiographic stage, knee pain severity, presence of bone marrow lesions, levels of biochemical markers of bone turnover (serum and/or urinary) and systemic bone mineral density status. Authors of original trials will be contacted to obtain individual patient data from each study. Risk of bias will be assessed using the Cochrane Collaboration's tool. The primary outcomes will include pain and radiographic joint space width loss. Studies using other MRI-based assessment of disease progression will also be eligible. Outcomes will be grouped into short-term (?3 months), intermediate-term (>3 months; ?12 months) and long-term (>12 months). Regression models will be used, adding an interaction term for each subgroup of interest to determine possible subgroup effects. There was no source of funding for this study. ETHICS AND DISSEMINATION:Dissemination of our findings is planned to occur through conference presentations, publication in peer-reviewed journals and social media. No formal ethics approval is generally required as no new data collection will be undertaken. PROSPERO REGISTRATION NUMBER:CRD42018093327.

Dataset Information

A siamese network with adaptive gated feature fusion for individual knee OA features grades prediction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets