No abstract

Background Cognitive frailty is the coexistence of physical frailty and cognitive impairment and is an at-risk state for many adverse health outcomes. Moderate-to-vigorous physical activity (MVPA) is protective against the progression of cognitive frailty. Physical inactivity is common in older people, and brisk walking is a feasible form of physical activity that can enhance their MVPA. Mobile health (mHealth) employing persuasive technology has been successful in increasing the levels of physical activity in older people. However, its feasibility and effects on older people with cognitive frailty are unclear. Objective We aimed to identify the issues related to the feasibility of an mHealth intervention and the trial (ie, recruitment, retention, participation, and compliance) and to examine the effects of the intervention on cognitive function, physical frailty, walking time, and MVPA. Methods An open-label, parallel design, randomized controlled trial (RCT) was employed. The eligibility criteria for the participants were age ≥60 years, having cognitive frailty, and having physical inactivity. In the intervention group, participants received both conventional behavior change intervention and mHealth (ie, smartphone-assisted program using Samsung Health and WhatsApp) interventions. In the control group, participants received conventional behavior change intervention only. The outcomes included cognitive function, frailty, walking time, and MVPA. Permuted block randomization in 1:1 ratio was used. The feasibility issue was described in terms of participant recruitment, retention, participation, and compliance. Wilcoxon signed-rank test was used to test the within-group effects in both groups separately. Results We recruited 99 participants; 33 eligible participants were randomized into either the intervention group (n=16) or the control (n=17) group. The median age was 71.0 years (IQR 9.0) and the majority of them were females (28/33, 85%). The recruitment rate was 33% (33/99), the participant retention rate was 91% (30/33), and the attendance rate of all the face-to-face sessions was 100% (33/33). The majority of the smartphone messages were read by the participants within 30 minutes (91/216, 42.1%). ActiGraph (58/66 days, 88%) and smartphone (54/56 days, 97%) wearing compliances were good. After the interventions, cognitive function improvement was significant in both the intervention (P=.003) and the control (P=.009) groups. The increase in frailty reduction (P=.005), walking time (P=.03), step count (P=.02), brisk walking time (P=.009), peak cadence (P=.003), and MVPA time (P=.02) were significant only in the intervention group. Conclusions Our mHealth intervention is feasible for implementation in older people with cognitive impairment and is effective at enhancing compliance with the brisk walking training program delivered by the conventional behavior change interventions. We provide preliminary evidence that this mHealth intervention can increase MVPA time to an extent sufficient to yield clinical benefits (ie, reduction in cognitive frailty). A full-powered and assessor-blinded RCT should be employed in the future to warrant these effects. Trial Registration HKU Clinical Trials Registry HKUCTR-2283; http://www.hkuctr.com/Study/Show/31df4708944944bd99e730d839db4756

Recognition of epileptic seizures from offline EEG signals is very important in clinical diagnosis of epilepsy. Compared with manual labeling of EEG signals by doctors, machine learning approaches can be faster and more consistent. However, the classification accuracy is usually not satisfactory for two main reasons: the distributions of the data used for training and testing may be different, and the amount of training data may not be enough. In addition, most machine learning approaches generate black-box models that are difficult to interpret. In this paper, we integrate transductive transfer learning, semi-supervised learning and TSK fuzzy system to tackle these three problems. More specifically, we use transfer learning to reduce the discrepancy in data distribution between the training and testing data, employ semi-supervised learning to use the unlabeled testing data to remedy the shortage of training data, and adopt TSK fuzzy system to increase model interpretability. Two learning algorithms are proposed to train the system. Our experimental results show that the proposed approaches can achieve better performance than many state-of-the-art seizure classification algorithms.

No abstract

Introduction: The objectives of this review paper were to synthesize the data from randomized controlled trials in the literature to come to a conclusion on the effects of e-health interventions on promoting physical activity in older people. Methods: The Medline, CINAHL, Embase, PsycINFO, and SportDiscus databases were searched for articles about studies that 1) recruited subjects with a mean age of > 50 years, 2) tested e-health interventions, 3) employed control groups with no or less advanced e-health strategies, 4) measured physical activity as an outcome, 5) were published between 1st January 2008 and 31st May 2019, and 6) employed randomized controlled trials. The risk of bias in individual studies was assessed using the Physiotherapy Evidence Database scale. To examine the effects of the interventions, variables quantifying the amount of physical activity were extracted. The within-group effects of individual studies were summarized using Hedges g and 95% confidence intervals. Between-group effects were summarized by meta-analyses using RevMan 5.0 with a random effect model. Results: Of the 2810 identified studies, 38 were eligible, 25 were included in the meta-analyses. The within-group effect sizes (Hedges g) of physical activity in the intervention group at T1 ranged from small to large: physical activity time (0.12 to 0.84), step counts (− 0.01 to 11.19), energy expenditure (− 0.05 to 0.86), walking time (0.13 to 3.33), and sedentary time (− 0.12 to − 0.28). The delayed effects as observed in T2 and T3 also ranged from small to large: physical activity time (0.24 to 1.24) and energy expenditure (0.15 to 1.32). In the meta-analysis, the betweengroup effect of the e-health intervention on physical activity time measured by questionnaires, physical activity time measured by objective wearable devices, energy expenditure, and step counts were all significant with minimal heterogeneity. Conclusion: E-health interventions are effective at increasing the time spent on physical activity, energy expenditure in physical activity, and the number of walking steps. It is recommended that e-health interventions be included in guidelines to enhance physical activity in older people. Further studies should be conducted to determine the most effective e-health strategies.

No abstract

No abstract

Classical fuzzy system modeling methods consider only the current scene where the training data are assumed to be fully collectable. However, if the data available from the current scene are insufficient, the fuzzy systems trained by using the incomplete datasets will suffer from weak generalization capability for the prediction in the scene. In order to overcome this problem, a knowledge-leverage-based fuzzy system (KL-FS) is studied in this paper from the perspective of transfer learning. The KL-FS intends to not only make full use of the data from the current scene in the learning procedure, but also effectively leverage the existing knowledge from the reference scenes. Specifically, a knowledge-leverage-based Takagi-Sugeno-Kang-type Fuzzy System (KL-TSK-FS) is proposed by integrating the corresponding knowledge-leverage mechanism. The new fuzzy system modeling technique is evaluated through experiments on synthetic and real-world datasets. The results demonstrate that KL-TSK-FS has better performance and adaptability than the traditional fuzzy modeling methods in scenes with insufficient data.