Automated osteoporosis classification and T-score prediction using hip radiographs via deep learning algorithm.

Background: Despite being the gold standard for diagnosing osteoporosis, dual-energy X-ray absorptiometry (DXA) is an underutilized screening tool for osteoporosis. Objectives: This study proposed and validated a controllable feature layer of a convolutional neural network (CNN) model with a preprocessing image algorithm to classify osteoporosis and predict T-score on the proximal hip region via simple hip radiographs. Design: This was a single-center, retrospective study. Methods: An image dataset of 3460 unilateral hip images from 1730 patients (age ⩾50 years) was retrospectively collected with matched DXA assessment for T-score for the targeted proximal hip regions to train (2473 unilateral hip images from 1430 patients) and test (497 unilateral hip images from 300 patients) the proposed CNN model. All images were processed with a fully automated CNN model, X1AI-Osteo. Results: The proposed screening tool illustrated a better performance (sensitivity: 97.2%; specificity: 95.6%; positive predictive value: 95.7%; negative predictive value: 97.1%; area under the curve: 0.96) than the open-sourced CNN models in predicting osteoporosis. Moreover, when combining variables, including age, body mass index, and sex as features in the training metric, there was high consistency in the T-score on the targeted hip regions between the proposed CNN model and the DXA (r = 0.996, p < 0.001). Conclusion: The proposed CNN model may identify osteoporosis and predict T-scores on the targeted hip regions from simple hip radiographs with high accuracy, highlighting the future application for population-based opportunistic osteoporosis screening with low cost and high adaptability for a broader population at risk. Trial registration: TMU-JIRB N201909036.

Validation of Gait Measurements on Short-Distance Walkways Using Azure Kinect DK in Patients Receiving Chronic Hemodialysis.

Muscle dysfunction, skeletal muscle fibrosis, and disability are associated with weakness in patients with end-stage renal disease. The main purpose of this study was to validate the effectiveness of a proposed system for gait monitoring on short-distance 1.5 m walkways in a dialysis center. Gaits with reduced speed and stride length, long sit-to-stand time (SST), two forward angles, and two unbalanced gait regions are defined in the proposed Kinect v3 gait measurement and analysis system (K3S) and have been considered clinical features in end-stage renal disease (ESRD) associated with poor dialysis outcomes. The stride and pace calibrations of the Kinect v3 system are based on the Zeno Walkway. Its single rating intraclass correlation (ICC) for the stride is 0.990, and its single rating ICC for the pace is 0.920. The SST calibration of Kinect v3 is based on a pressure insole; its single rating ICC for the SST is 0.871. A total of 75 patients on chronic dialysis underwent gait measurement and analysis during walking and weighing actions. After dialysis, patients demonstrated a smaller stride (p < 0.001) and longer SST (p < 0.001). The results demonstrate that patients' physical fitness was greatly reduced after dialysis. This study ensures patients' adequate physical gait strength to cope with the dialysis-associated physical exhaustion risk by tracing gait outliers. As decreased stride and pace are associated with an increased risk of falls, further studies are warranted to evaluate the clinical benefits of monitoring gait with the proposed reliable and valid system in order to reduce fall risk in hemodialysis patients.

Association between hyperlipidemia and trigger finger: A nationwide population-based cohort study.

The cause of trigger fingers remains uncertain. High lipid levels in the blood may reduce blood supply to the distal fingers and promote inflammation. We aimed to explore the association between hyperlipidemia and trigger finger. A nationwide population-based cohort study using longitudinal data from 2000 to 2013, 41,421 patients were included in the hyperlipidemia cohort and 82,842 age- and sex-matched patients were included in the control cohort. The mean age was 49.90 ± 14.73 years in the hyperlipidemia cohort and 49.79 ± 14.71 years in the control cohort. After adjusting for possible comorbidities, the hazard ratio of trigger finger in the hyperlipidemia cohort was 4.03 (95% confidence interval [CI], 3.57-4.55), with values of 4.59 (95% CI, 3.67-5.73) and 3.77 (95% CI, 3.26-4.36) among male and female patients, respectively. This large-scale population-based study demonstrated that hyperlipidemia is correlated to trigger finger.

Long-term ambient hydrocarbon exposure and incidence of urinary bladder cancer.

Particulate matter and volatile organic compounds, including total hydrocarbons (THCs), are major ambient air pollutants. Primary nonmethane hydrocarbons (NMHCs) originate from vehicle emissions. The association between air pollution and urinary bladder cancer (UBC) is debatable. We investigated whether long-term exposure to ambient hydrocarbons increases UBC risk among people aged ≥ 20 years in Taiwan. Linkage dataset research with longitudinal design was conducted among 589,135 initially cancer-free individuals during 2000-2013; 12 airborne pollutants were identified. Several Cox models considering potential confounders were employed. The study outcomes were invasive or in situ UBC incidence over time. The targeted pollutant concentration was divided into three tertiles: T1/T2/T3. The mean age of individuals at risk was 42.5 (SD 15.7), and 50.5% of the individuals were men. The mean daily average over 10 years of airborne THC concentration was 2.25 ppm (SD 0.13), and NMHC was 0.29 ppm (SD 0.09). Both pollutants show long-term monotonic downward trend over time using the Mann-Kendall test. There was a dose-dependent increase in UBC at follow-up. UBC incidence per 100,000 enrollees according to T1/T2/T3 exposure to THC was 60.9, 221.2, and 651.8, respectively; it was 170.0/349.5/426.7 per 100,000 enrollees, corresponding to T1/T2/T3 exposure to NMHC, respectively. Without controlling for confounding air pollutants, the adjusted hazard ratio (adj.HR) was 1.83 (95% CI 1.75-1.91) per 0.13-ppm increase in THC; after controlling for PM2.5, adj.HR was even higher at 2.09 (95% CI 1.99-2.19). The adj.HR was 1.37 (95% CI 1.32-1.43) per 0.09-ppm increase in ambient NMHC concentration. After controlling for SO2 and CH4, the adj.HR was 1.10 (95% CI 1.06-1.15). Sensitivity analyses showed that UBC development risk was not sex-specific or influenced by diabetes status. Long-term exposure to THC and NMHC may be a risk factor for UBC development. Acknowledging pollutant sources can inform risk management strategies.

Enhanced Risk of Osteoporotic Fracture in Patients with Sarcopenia: A National Population-Based Study in Taiwan.

Sarcopenia is a progressive and generalized skeletal muscle disorder associated with poor health outcomes in older adults. However, its association with the risk of fracture risk is yet to be clarified. Therefore, this study aimed to assess the incidence and consequence of osteoporosis-related fractures among patients with sarcopenia in Taiwan. A retrospective, population-based study on 616 patients with sarcopenia, aged >40 years, and 1232 individuals without sarcopenia was conducted to evaluate claims data from Taiwan's National Health Insurance Research Database collected in the period January 2000−December 2013. The incidence rate of osteoporosis-related fracture was 18.13 and 14.61 per 1000 person years in the patients with sarcopenia and comparison cohort, respectively. Patients with sarcopenia had a greater osteoporotic fracture risk (adjusted hazard ratio [HR] 2.11; 95% confidence interval [CI] 1.47−3.04) after correcting for possible confounding. Additionally, females showed statistically significant correlations of sarcopenia with osteoporosis-related fracture risk (HR 1.53; CI 0.83−2.8 for males and HR 2.40, CI 1.51−3.81 for females). During this retrospective study on the fracture risk in Taiwan, an adverse impact of sarcopenia was observed, which substantiates the need to work toward sarcopenia prevention and interventions to reverse fracture susceptibility in patients with sarcopenia.

Late-onset epilepsy and subsequent increased risk of dementia.

Inflammation is considered as a key pathogenesis factor of dementia and epilepsy. However, epilepsy's association with dementia, particularly its role in the development of dementia, remains unclear. To evaluate the association between epilepsy and the risk of dementia, in Taiwan, we have now conducted a retrospective cohort study comprising 675 individuals (age, ≥50 years) with epilepsy and 2,025 matched control subjects without epilepsy. In order to match individuals diagnosed with epilepsy with those with no diagnosis of epilepsy (comparison cohort), we utilized exact matching at a ratio of 1:3. Compared with those in the comparison cohort, individuals in the epilepsy cohort had a significantly increased risk of developing dementia (adjusted hazard ratio = 2.87, p < 0.001). A similar result has been observed after stratifying for sex (adjusted hazard ratio in males = 2.95, p < 0.001; adjusted hazard ratio in females = 2.66, p < 0.001). To conclude, based on these data, epileptic individuals ≥50 years were at a greater risk of developing dementia than people who do not have epilepsy, which indicates that a diagnosis of epilepsy presents a greater risk for the development of dementia.

Robustness design of fuzzy control for nonlinear multiple time-delay large-scale systems via neural-network-based approach.

The stabilization problem is considered in this correspondence for a nonlinear multiple time-delay large-scale system. First, the neural-network (NN) model is employed to approximate each subsystem. Then, a linear differential inclusion (LDI) state-space representation is established for the dynamics of each NN model. According to the LDI state-space representation, a robustness design of fuzzy control is proposed to overcome the effect of modeling errors between subsystems and NN models. Next, in terms of Lyapunov's direct method, a delay-dependent stability criterion is derived to guarantee the asymptotic stability of nonlinear multiple time-delay large-scale systems. Finally, based on this criterion and the decentralized control scheme, a set of fuzzy controllers is synthesized to stabilize the nonlinear multiple time-delay large-scale system.