Perception is everything: OA is exciting

DC-AAE: Dual channel adversarial autoencoder with multitask learning for KL-grade classification in knee radiographs

Knee osteoarthritis (OA) is a frequent musculoskeletal disorder that leads to physical disability in older adults. Manual OA assessment is performed via visual inspection, which is highly subjective as it suffers from moderate to high inter-observer variability. Many deep learning-based techniques have been proposed to address this issue. However, owing to the limited amount of labelled data, all existing solutions have limitations in terms of performance or the number of classes. This paper proposes a novel fully automatic Kellgren and Lawrence (KL) grade classification scheme in knee radiographs. We developed a semi-supervised multi-task learning-based approach that enables the exploitation of additional unlabelled data in an unsupervised as well as supervised manner. Specifically, we propose a dual-channel adversarial autoencoder, which is first trained in an unsupervised manner for reconstruction tasks only. To exploit the additional data in a supervised way, we propose a multi-task learning framework by introducing an auxiliary task. In particular, we use leg side identification as an auxiliary task, which allows the use of more datasets, e.g., CHECK dataset. The work demonstrates that the utilization of additional data can improve the primary task of KL-grade classification for which only limited labelled data is available. This semi-supervised learning essentially helps to improve the feature learning ability of our framework, which leads to improved performance for KL-grade classification. We rigorously evaluated our proposed model on the two largest publicly available datasets for various aspects, i.e., overall performance, the effect of additional unlabelled samples and auxiliary tasks, robustness analysis, and ablation study. The proposed model achieved the accuracy, precision, recall, and F1 score of 75.53%, 74.1%, 78.51%, and 75.34%, respectively. Furthermore, the experimental results show that the suggested model not only achieves state-of-the-art performance on two publicly available datasets but also exhibits remarkable robustness.

A joint multi-modal learning method for early-stage knee osteoarthritis disease classification

Osteoarthritis (OA) is a progressive and chronic disease. Identifying the early stages of OA disease is important for the treatment and care of patients. However, most state-of-the-art methods only use single-modal data to predict disease status, so that these methods usually ignore complementary information in multi-modal data. In this study, we develop an integrated multi-modal learning method (MMLM) that uses an interpretable strategy to select and fuse clinical, imaging, and demographic features to classify the grade of early-stage knee OA disease. MMLM applies XGboost and ResNet50 to extract two heterogeneous features from the clinical data and imaging data, respectively. And then we integrate these extracted features with demographic data. To avoid the negative effects of redundant features in a direct integration of multiple features, we propose a L1-norm-based optimization method (MMLM) to regularize the inter-correlations among the multiple features. MMLM was assessed using the Osteoarthritis Initiative (OAI) data set with machine learning classifiers. Extensive experiments demonstrate that MMLM improves the performance of the classifiers. Furthermore, a visual analysis of the important features in the multimodal data verified the relations among the modalities when classifying the grade of knee OA disease.

Untacted automated robotic upper-trunk- lower reciprocal locomotor training for knee osteoarthritis: A randomized controlled trial

BACKGROUND

Although millions of people with osteoarthritis (OA) have altered biomechanical alignment, movement, and knee joint pain during gait, there are no effective and sustainable interventions. To mitigate such impairments, we developed an untacted self-automated robotic and electromyography (EMG)-augmented upper-trunk-lower reciprocal locomotor training (SRGT) intervention.

OBJECTIVE

To compare the effects of SRGT and conventional treadmill gait training (CTGT) on the medial knee joint space width (JSW), hip adduction moment (HAM), knee varus deformity, pain, and physical function in community-dwelling older adults with OA.

METHODS

Older adults diagnosed with medial compartment knee OA (5 men, 35 women; mean age = 78.50 ± 9.10 years) were recruited and underwent either SRGT or CTGT, 30 min a day, 3 times a week, over a 4-week period. Outcome measurements included the JSW, HAM, knee varus angle (VA), and Western Ontario McMaster Universities osteoarthritis index (WOMAC).

RESULTS

Analysis of covariance (ANCOVA) showed that SRGT ed to greater changes in medial knee JSW (p= 0.00001), HAM (p= 0.00001), VA (p= 0.00001), and WOMAC (p= 0.00001) scores.

CONCLUSION

This study provides the first evidence for the long-term clinical and biomechanical effects of SRGT on JSW, knee joint kinematics, kinetics, and WOMAC scores in older adults with OA. Most importantly, self-automatic robotic gait training may be an alternative, effective, and sustainable treatment for the upper-trunk-lower reciprocal locomotor training in older adults with OA.

Discriminative Regularized Auto-Encoder for Early Detection of Knee OsteoArthritis: Data from the Osteoarthritis Initiative

OsteoArthritis (OA) is the most common disorder of the musculoskeletal system and the major cause of reduced mobility among seniors. The visual evaluation of OA still suffers from subjectivity. Recently, Computer-Aided Diagnosis (CAD) systems based on learning methods showed potential for improving knee OA diagnostic accuracy. However, learning discriminative properties can be a challenging task, particularly when dealing with complex data such as X-ray images, typically used for knee OA diagnosis. In this paper, we introduce a Discriminative Regularized Auto Encoder (DRAE) that allows to learn both relevant and discriminative properties that improve the classification performance. More specifically, a penalty term, called discriminative loss is combined with the standard Auto-Encoder training criterion. This additional term aims to force the learned representation to contain discriminative information. Our experimental results on data from the public multicenter OsteoArthritis Initiative (OAI) show that the developed method presents potential results for early knee OA detection.

Consumption of dairy products in relation to the presence of clinical knee osteoarthritis: The Maastricht Study

PURPOSE

Observational studies showed inverse associations between milk consumption and knee osteoarthritis (knee OA). There is lack of information on the role of specific dairy product categories. We explored the association between dairy consumption and the presence of knee osteoarthritis in 3010 individuals aged 40-75 years participating in The Maastricht Study.

METHODS

The presence of knee OA was defined according to a slightly modified version of the American College of Rheumatology (ACR) clinical classification criteria. Data on dairy consumption were appraised by a 253-item FFQ covering 47 dairy products with categorization on fat content, fermentation or dairy type. Multivariable logistic regression analyses were performed to estimate odd ratios (ORs) and 95% confidence intervals (95%CI), while correcting for relevant factors.

RESULTS

427 (14%) participants were classified as having knee OA. Significant inverse associations were observed between the presence of knee OA and intake of full-fat dairy and Dutch, primarily semi-hard, cheese, with OR for the highest compared to the lowest tertile of intake of 0.68 (95%CI 0.50-0.92) for full-fat dairy, and 0.75 (95%CI 0.56-0.99) for Dutch cheese. No significant associations were found for other dairy product categories.

CONCLUSION

In this Dutch population, higher intake of full-fat dairy and Dutch cheese, but not milk, was cross-sectionally associated with the lower presence of knee OA. Prospective studies need to assess the relationship between dairy consumption, and in particular semi-hard cheeses, with incident knee OA.

Open-Loop Audiovisual Stimulation Induces Delta EEG Activity in Older Adults With Osteoarthritis Pain and Insomnia

OBJECTIVE

People with chronic insomnia tend to have cortical hyperarousal marked by excessive beta-/gamma-frequency brain activity during both wake and sleep. Currently, treatment options for managing hyperarousal are limited. Open-loop audiovisual stimulation (AVS) may be such a treatment. The purpose of this study was to provide a mechanistic foundation for future AVS research in sleep promotion by examining quantitative electroencephalogram (QEEG) responses to an AVS sleep-induction program.

METHOD

Sixteen older adults with both chronic insomnia and osteoarthritis pain were randomly assigned to either active- or placebo-control AVS. Electroencephalogram (EEG) was collected during baseline (5 min, eyes closed/resting) and throughout 30 min of AVS.

RESULTS

Findings showed significantly elevated mean baseline gamma (35-45 Hz) power in both groups compared to an age- and gender-matched, noninsomnia normative database, supporting cortical hyperarousal. After 30 min of exposure to AVS, the active group showed significantly increased delta power compared to the placebo-control group, providing the first controlled evidence that active AVS induction increases delta QEEG activity in insomnia patients and that these changes are immediate. In the active group, brain locations that showed the most delta induction (Cz, Fp, O1, and O2) were associated with the sensory-thalamic pathway, consistent with the sensory stimulation provided by the active AVS program.

CONCLUSIONS

Findings demonstrate that delta induction, which can promote sleep, is achievable using a 30-min open-loop AVS program. The potential for AVS treatment of insomnia in the general population remains to be demonstrated in well-designed clinical trials.

Single Molecule Microscopy Reveals an Increased Hyaluronan Diffusion Rate in Synovial Fluid from Knees Affected by Osteoarthritis

Osteoarthritis is a common and progressive joint disorder. Despite its widespread, in clinical practice only late phases of osteoarthritis that are characterized by severe joint damage are routinely detected. Since osteoarthritis cannot be cured but relatively well managed, an early diagnosis and thereby early onset of disease management would lower the burden of osteoarthritis. Here we evaluated if biophysical parameters of small synovial fluid samples extracted by single molecule microscopy can be linked to joint damage. In healthy synovial fluid (ICRS-score < 1) hyaluronan showed a slower diffusion (2.2 μm²/s, N = 5) than in samples from patients with joint damage (ICRS-score > 2) (4.5 μm²/s, N = 16). More strikingly, the diffusion coefficient of hyaluronan in healthy synovial fluid was on average 30% slower than expected by sample viscosity. This effect was diminished or missing in samples from patients with joint damage. Since single molecule microscopy needs only microliters of synovial fluid to extract the viscosity and the specific diffusion coefficient of hyaluronan this method could be of use as diagnostic tool for osteoarthritis.

Improving musculoskeletal health: global issues

Musculoskeletal (MSK) disorders are among the leading reasons why patients consult a family or primary health practitioner, take time off work and become disabled. Many of the MSK disorders are more common in the elderly. Thus, as the proportion of the elderly increases all over the world, MSK disorders will make a greater contribution to the global burden of disease. Epidemiological studies have shown that the spectrum of MSK disorders in developing countries is similar to that seen in industrialised countries, but the burden of disease tends to be higher due to a delay in diagnosis or lack of access to adequate health-care facilities for effective treatment. Musculoskeletal pain is very common in the community while fibromyalgia is being recognised as part of a continuum of chronic widespread pain rather than a narrowly defined entity. This will allow research to improve our understanding of pain in a variety of diffuse pain syndromes. The availability of newer more effective therapies has resulted in efforts to initiate therapy at an earlier stage of diseases. The new criteria for rheumatoid arthritis, and the diagnosis of axial and peripheral involvement in spondyloarthritis, permit an earlier diagnosis without having to wait for radiological changes. One of the major health challenges is the global shortage of health workers, and based on current training of health workers and traditional models of care for service delivery, the global situation is unlikely to change in the near future. Thus, new models of care and strategies to train community health-care workers and primary health-care practitioners to detect and initiate the management of patients with MSK disorders at an earlier stage are required. There is also a need for prevention strategies with campaigns to educate and raise awareness among the entire population. Lifestyle interventions such as maintaining an ideal body weight to prevent obesity, regular exercises, avoidance of smoking and alcohol abuse, intake of a balanced diet and nutrients to include adequate calcium and vitamin D, modification of the work environment and avoidance of certain repetitive activities will prevent or ameliorate disorders such as osteoarthritis, osteoporosis, rheumatoid arthritis, gout and MSK pain syndromes including low back pain and work-related pain syndromes. These prevention strategies also contribute to reducing the prevalence and outcome of diseases such as hypertension, cardiovascular diseases, diabetes and respiratory diseases. Thus, prevention strategies require urgent attention globally.