Cross-subject EMG hand gesture recognition based on dynamic domain generalization.

Electromyography (EMG) signal based cross-subject gesture recognition methods reduce the influence of individual differences using transfer learning technology. These methods generally require calibration data collected from new subjects to adapt the pre-trained model to existing subjects. However, collecting calibration data is usually trivial and inconvenient for new subjects. This is currently a major obstacle to the daily use of hand gesture recognition based on EMG signals. To tackle the problem, we propose a novel dynamic domain generalization (DDG) method which is able to achieve accurate recognition on the hand gesture of new subjects without any calibration data. In order to extract more robust and adaptable features, a meta-adjuster is leveraged to generate a series of template coefficients to dynamically adjust dynamic network parameters. Specifically, two different kinds of templates are designed, in which the first one is different kinds of features, such as temporal features, spatial features, and spatial-temporal features, and the second one is different normalization layers. Meanwhile, a mix-style data augmentation method is introduced to make the meta-adjuster's training data more diversified. Experimental results on a public dataset verify that the proposed DDG outperforms the counterpart methods.

The new checklist to evaluate educational value of YouTube surgical videos of transurethral resection of the prostate.

Background: Transurethral resection of the prostate (TURP) is a widespread, effective way to treat benign prostatic hyperplasia (BPH). Many medical students and junior clinicians increasingly turn to easily accessible online resources to learn this technique, such as videos on YouTube. This study assessed the educational value of YouTube videos about TURP, which are popular among many young surgeons. Methods: We searched YouTube as of August 2, 2022 for videos fulfilling the search terms "transurethral resection of the prostate", "benign prostatic hyperplasia", "BPH", "TURP", "benign prostatic enlargement", "bladder outlet obstruction" and "lower urinary tract symptom". We assessed the educational value of the identified videos using a custom-designed checklist. Results: We identified 47 relevant videos, 20 of which were posted after July 1, 2020. The average number of views was 576,379±208,535 (range, 54-1,385,713). The average quality score of the videos was 7.38±2.53 (range, 4-12) on a 15-point scale, and 20 were judged to be of low educational quality. Quality scores correlated positively with the number of likes (R=0.596, P<0.01). Conclusions: The educational value of most TURP videos on YouTube appears to be low, with most lacking detailed explanations of preoperative preparations and the surgical procedure. High-quality video resources about TURP need to be developed for medical students and junior surgeons. Standard quality criteria should also be developed and disseminated to ensure the production of accurate learning resources for junior clinicians.

Vibroarthrography-based Knee Lesions Location via Multi-Label Embedding Learning.

Vibration arthrography (VAG) signals are widely utilized for knee pathology recognition due to their non-invasive and radiation-free nature. While most studies focus on determining knee health status, few have examined using VAG signals to locate knee lesions, which would greatly aid physicians in diagnosis and patient monitoring. To address this, we propose using Multi-Label classification (MLC) to efficiently locate different types of lesions within a single input. However, current MLC methods are not suitable for knee lesion location due to two major issues: 1) the positive-negative imbalance of pathological labels in knee pathology recognition is not considered, leading to poor performance, and 2) sparse label correlations between different lesions cannot be effectively extracted. Our solution is a label autoencoder incorporating a pre-trained model (PTM-LAE). To mitigate the positive-negative disequilibrium, we propose a pre-trained feature mapping model utilizing focal loss to dynamically adjust sample weights and focus on difficult-to-classify samples. To better explore the correlations between sparse labels, we introduce a Factorization-Machine-based neural network (DeepFM) that combines higher-order and lower-order correlations between different lesions. Experiments on our collected VAG data demonstrate that our model outperforms state-of-the-art methods.

Association Between Urinary Bisphenols and Body Composition Among American Adults: Cross-Sectional National Health and Nutrition Examination Survey Study.

BACKGROUND: Bisphenol A (BPA), bisphenol S (BPS), and bisphenol F (BPF) are widely used in various consumer products. They are environmental contaminants with estrogenic properties that have been linked to various health outcomes. Understanding their impact on body composition is crucial for identifying potential health risks and developing preventive strategies. However, most current studies have only focused on their relationship with BMI. OBJECTIVE: This study aimed to investigate the association between urinary levels of BPA, BPS, and BPF and body composition, including BMI, lean mass, and fat mass, in a large population-based sample. METHODS: We conducted a cross-sectional analysis using data from the National Health and Nutrition Examination Survey 2003-2016. Body composition data were assessed using dual-energy X-ray absorptiometry, which provided precise measurements of lean mass, fat mass, and other indicators. We used multivariate linear regression models to estimate the associations, adjusting for potential confounders such as age, gender, race, socioeconomic factors, and lifestyle variables. RESULTS: The results revealed significant associations between bisphenol exposure and body composition. After adjusting for covariates, BPS showed a positive association with BMI, with quartiles 3 and 4 having 0.91 (95% CI 0.34-1.48) and 1.15 (95% CI 0.55-1.74) higher BMI, respectively, compared with quartile 1 (P<.001). BPA was negatively associated with total lean mass (TLM) and appendicular lean mass, with quartiles 2, 3, and 4 having -7.85 (95% CI -11.44 to -4.25), -12.33 (95% CI -16.12 to -8.54), and -11.08 (95% CI -15.16 to -7.01) lower TLM, respectively, compared with quartile 1 (P<.001). BPS was negatively associated with TLM, with quartiles 3 (ß=-10.53, 95% CI -16.98 to -4.08) and 4 (ß=-11.14, 95% CI -17.83 to -4.45) having significantly lower TLM (P=.005). Both BPA and BPS showed a positive dose-response relationship with trunk fat (BPA: P=.002; BPS: P<.001) and total fat (BPA: P<.001; BPS: P=.01). No significant association was found between BPF and any body composition parameter. CONCLUSIONS: This large-sample study highlights the associations between urinary levels of BPA and BPS and alterations in body composition, including changes in lean mass, fat mass, and regional fat distribution. These findings underscore the importance of understanding the potential health risks associated with bisphenol exposure and emphasize the need for targeted interventions to mitigate adverse effects on body composition.