Real-Time Anomaly Detection for an ADMM-Based Optimal Transmission Frequency Management System for IoT Devices.

In this paper, we investigate different scenarios of anomaly detection on decentralised Internet of Things (IoT) applications. Specifically, an anomaly detector is devised to detect different types of anomalies for an IoT data management system, based on the decentralised alternating direction method of multipliers (ADMM), which was proposed in our previous work. The anomaly detector only requires limited information from the IoT system, and can be operated using both a mathematical-rule-based approach and the deep learning approach proposed in the paper. Our experimental results show that detection based on mathematical approach is simple to implement, but it also comes with lower detection accuracy (78.88%). In contrast, the deep-learning-enabled approach can easily achieve a higher detection accuracy (96.28%) in the real world working environment.

Exploring the relationships between fundamental movement skills and health related fitness components in children.

This study examined the relationship between fundamental movement skills (FMS) and health related fitness (HRF) components in children. A cross section of Irish primary school children across all age groups participated in this study (n=2098, 47% girls, age 5-12 years of age, mean age 9.2 ± 2.04). FMS were measured using the Test of Gross Motor Development (TGMD-3), along with two additional assessments of vertical jump and balance. All HRF components were also assessed: body composition through BMI and waist circumference, muscular strength (MS) using a hand dynamometer, muscular endurance (ME) through the plank test, flexibility with back-saver sit-and-reach, and cardiovascular endurance (CVE) using the 20 m PACER test. Hierarchal multiple regressions were used to measure associations between the HRF components and overall FMS and the FMS subtests: locomotor, object control and balance skills. Results show significant positive relationships between FMS and MS (R2 = 0.25, ß= -0.19), ME (R2 = 0.11, ß = 0.34), flexibility (R2 = 0.13, ß = 0.14) and CVE (R2 = 0.17, ß = 0.39), and an inverse relationship between FMS and body composition (R2 = 0.25, ß= -0.19). The data presented reinforces the position that the relationship between FMS and HRF is dynamic, and predominantly strengthens with age through the course of childhood. Findings suggest that developing FMS as a child may be important to developing HRF across childhood and into adolescence.

TGMD-3 short version: Evidence of validity and associations with sex in Irish children.

This study examined the internal structure and evidence of validity of the Test of Gross Motor Development 3rd edition (TGMD-3) in primary school aged children. Participants (n = 1608, 47% girls, age range 5-11 years, mean age 9.2 ± 2.04) were recruited from Irish schools across twelve counties (56% rural, 44% urban). The TGMD-3 was used to measure FMS proficiency (Ulrich, 2020). A two-factor model (13 skills) was used and confirmatory indexes were calculated. The Bayesian criteria and the Composite Reliability were employed to evaluate alternative models. Relationships between the final model proposed with age, sex and BMI were calculated using a network analysis. Mplus 8.0 and Rstudio were used. A two-factor model (locomotion and object control) with adequate values (> 0.30) for the seven skills (gallop, hop, jump, two-hand strike, bounce, catch, overhand throw) presented excellent indexes. The skills with the highest indicator of strength centrality in the network were bounce and catch for both boys and girls and hop for boys and horizontal jump for girls. This study evidences the validity and reliability of the internal structure of the TGMD-3 and demonstrates that a short version of the TGMD-3, comprising seven skills is a valid measure of FMS in this population.

Recognition and Repetition Counting for Local Muscular Endurance Exercises in Exercise-Based Rehabilitation: A Comparative Study Using Artificial Intelligence Models.

Exercise-based cardiac rehabilitation requires patients to perform a set of certain prescribed exercises a specific number of times. Local muscular endurance exercises are an important part of the rehabilitation program. Automatic exercise recognition and repetition counting, from wearable sensor data, is an important technology to enable patients to perform exercises independently in remote settings, e.g., their own home. In this paper, we first report on a comparison of traditional approaches to exercise recognition and repetition counting (supervised ML and peak detection) with Convolutional Neural Networks (CNNs). We investigated CNN models based on the AlexNet architecture and found that the performance was better than the traditional approaches, for exercise recognition (overall F1-score of 97.18%) and repetition counting (±1 error among 90% observed sets). To the best of our knowledge, our approach of using a single CNN method for both recognition and repetition counting is novel. Also, we make the INSIGHT-LME dataset publicly available to encourage further research.

Bio-inspired Surface Texture Modification as a Viable Feature of Future Aquatic Antifouling Strategies: A Review.

The imitation of natural systems to produce effective antifouling materials is often referred to as "biomimetics". The world of biomimetics is a multidisciplinary one, needing careful understanding of "biological structures", processes and principles of various organisms found in nature and based on this, designing nanodevices and nanomaterials that are of commercial interest to industry. Looking to the marine environment for bioinspired surfaces offers researchers a wealth of topographies to explore. Particular attention has been given to the evaluation of textures based on marine organisms tested in either the laboratory or the field. The findings of the review relate to the numbers of studies on textured surfaces demonstrating antifouling potential which are significant. However, many of these are only tested in the laboratory, where it is acknowledged a very different response to fouling is observed.

A Quality of Experience assessment of haptic and augmented reality feedback modalities in a gait analysis system.

Gait analysis is a technique that is used to understand movement patterns and, in some cases, to inform the development of rehabilitation protocols. Traditional rehabilitation approaches have relied on expert guided feedback in clinical settings. Such efforts require the presence of an expert to inform the re-training (to evaluate any improvement) and the patient to travel to the clinic. Nowadays, potential opportunities exist to employ the use of digitized "feedback" modalities to help a user to "understand" improved gait technique. This is important as clear and concise feedback can enhance the quality of rehabilitation and recovery. A critical requirement emerges to consider the quality of feedback from the user perspective i.e. how they process, understand and react to the feedback. In this context, this paper reports the results of a Quality of Experience (QoE) evaluation of two feedback modalities: Augmented Reality (AR) and Haptic, employed as part of an overall gait analysis system. The aim of the feedback is to reduce varus/valgus misalignments, which can cause serious orthopedics problems. The QoE analysis considers objective (improvement in knee alignment) and subjective (questionnaire responses) user metrics in 26 participants, as part of a within subject design. Participants answered 12 questions on QoE aspects such as utility, usability, interaction and immersion of the feedback modalities via post-test reporting. In addition, objective metrics of participant performance (angles and alignment) were also considered as indicators of the utility of each feedback modality. The findings show statistically significant higher QoE ratings for AR feedback. Also, the number of knee misalignments was reduced after users experienced AR feedback (35% improvement with AR feedback relative to baseline when compared to haptic). Gender analysis showed significant differences in performance for number of misalignments and time to correct valgus misalignment (for males when they experienced AR feedback). The female group self-reported higher utility and QoE ratings for AR when compared to male group.

Assistive technology: Understanding the needs and experiences of individuals with autism spectrum disorder and/or intellectual disability in Ireland and the UK.

Assistive technologies (ATs) aimed at improving the life quality of persons with Autism Spectrum Disorder and/or Intellectual Disability (ASD/ID) is an important research area. Few have examined how this population use and experience AT or their vision for future uses of AT. The present study aimed to update and extend previous research and provides insight from caregivers, and other stakeholders (n = 96), living in Ireland and the United Kingdom, on their experiences of assistive technology (AT) for ASD/ID. Caregiver and professional responses to an anonymous online survey showed that focus individuals were rated low in terms of independent and self-management skills, with scheduling and planning and communication identified as desirable future AT functions. Overall, positive experiences of AT were reported, with AT use more than doubling in recent years.

Moving Well-Being Well: Investigating the maturation of fundamental movement skill proficiency across sex in Irish children aged five to twelve.

Fundamental movement skills (FMS) are the basic building blocks of more advanced, complex movements required to participate in physical activity. This study examined FMS proficiency across the full range of Irish primary school children (n = 2098, 47% girls, age range 5-12 years). Participants were assessed using the Test of Gross Motor Development, 3rd edition (TGMD-3), Victorian Fundamental Movement skills manual, and the balance subtest from the Bruininks-Oseretsky Test of Motor Proficiency 2 (BOT-2). Independent sample t-tests and a one way between groups ANOVA with planned comparisons were used analyse sex and age differences. Mastery or near mastery of skills ranged from 16% for overhand throw, to 75.3% for run. Girls scored significantly higher than boys in the locomotor and balance subtests with the boys outperforming the girls in object control skills. Improvements in ability can be seen over time (F(8,1968) = 70.18, p < 0.001), with significant increases in FMS proficiency seen up to the age of 10, after which proficiency begins to decline. The findings demonstrate the low levels of FMS proficiency amongst Irish primary school children, the differences between sex that exist, and highlights the need for more programmes that focus on developing these FMS at an early age.

A Low-Cost Smart Sensor Network for Catchment Monitoring.

Understanding hydrological processes in large, open areas, such as catchments, and further modelling these processes are still open research questions. The system proposed in this work provides an automatic end-to-end pipeline from data collection to information extraction that can potentially assist hydrologists to better understand the hydrological processes using a data-driven approach. In this work, the performance of a low-cost off-the-shelf self contained sensor unit, which was originally designed and used to monitor liquid levels, such as AdBlue, fuel, lubricants etc., in a sealed tank environment, is first examined. This process validates that the sensor does provide accurate water level information for open water level monitoring tasks. Utilising the dataset collected from eight sensor units, an end-to-end pipeline of automating the data collection, data processing and information extraction processes is proposed. Within the pipeline, a data-driven anomaly detection method that automatically extracts rapid changes in measurement trends at a catchment scale. The lag-time of the test site (Dodder catchment Dublin, Ireland) is also analyzed. Subsequently, the water level response in the catchment due to storm events during the 27 month deployment period is illustrated. To support reproducible and collaborative research, the collected dataset and the source code of this work will be publicly available for research purposes.

Predicting knee osteoarthritis severity: comparative modeling based on patient's data and plain X-ray images.

Knee osteoarthritis (KOA) is a disease that impairs knee function and causes pain. A radiologist reviews knee X-ray images and grades the severity level of the impairments according to the Kellgren and Lawrence grading scheme; a five-point ordinal scale (0-4). In this study, we used Elastic Net (EN) and Random Forests (RF) to build predictive models using patient assessment data (i.e. signs and symptoms of both knees and medication use) and a convolution neural network (CNN) trained using X-ray images only. Linear mixed effect models (LMM) were used to model the within subject correlation between the two knees. The root mean squared error for the CNN, EN, and RF models was 0.77, 0.97 and 0.94 respectively. The LMM shows similar overall prediction accuracy as the EN regression but correctly accounted for the hierarchical structure of the data resulting in more reliable inference. Useful explanatory variables were identified that could be used for patient monitoring before X-ray imaging. Our analyses suggest that the models trained for predicting the KOA severity levels achieve comparable results when modeling X-ray images and patient data. The subjectivity in the KL grade is still a primary concern.

Few-shot hypercolumn-based mitochondria segmentation in cardiac and outer hair cells in focused ion beam-scanning electron microscopy (FIB-SEM) data.

We present a novel AI-based approach to the few-shot automated segmentation of mitochondria in large-scale electron microscopy images. Our framework leverages convolutional features from a pre-trained deep multilayer convolutional neural network, such as VGG-16. We then train a binary gradient boosting classifier on the resulting high-dimensional feature hypercolumns. We extract VGG-16 features from the first four convolutional blocks and apply bilinear upsampling to resize the obtained maps to the input image size. This procedure yields a 2688-dimensional feature hypercolumn for each pixel in a 224 × 224 input image. We then apply L 1-regularized logistic regression for supervised active feature selection to reduce dependencies among the features, to reduce overfitting, as well as to speed-up gradient boosting-based training. During inference we block process 1728 × 2022 large microscopy images. Our experiments show that in such a formulation of transfer learning our processing pipeline is able to achieve high-accuracy results on very challenging datasets containing a large number of irregularly shaped mitochondria in cardiac and outer hair cells. Our proposed few-shot training approach gives competitive performance with the state-of-the-art using far less training data.

Real-time field sports scene classification using colour and frequency space decompositions.

This paper presents a novel approach to recognize a scene presented in an image with specific application to scene classification in field sports video. We propose different variants of the algorithm ranging from bags of visual words to the simplified real-time implementation, that takes only the most important areas of similar colour into account. All the variants feature similar accuracy which is comparable to very well-known image indexing techniques like SIFT or HoGs. For the comparison purposes, we also developed a specific database which is now available online. The algorithm is suitable in scene recognition task thanks to changes in speed and robustness to the image resolution, thus, making it a good candidate in real-time video indexing systems. The procedure features high simplicity thanks to the fact that it is based on the very well-known Fourier transform.

Behavior Change Techniques in Physical Activity eHealth Interventions for People With Cardiovascular Disease: Systematic Review.

BACKGROUND: Cardiovascular disease (CVD) is the leading cause of premature death and disability in Europe, accounting for 4 million deaths per year and costing the European Union economy almost €196 billion annually. There is strong evidence to suggest that exercise-based secondary rehabilitation programs can decrease the mortality risk and improve health among patients with CVD. Theory-informed use of behavior change techniques (BCTs) is important in the design of cardiac rehabilitation programs aimed at changing cardiovascular risk factors. Electronic health (eHealth) is the use of information and communication technologies (ICTs) for health. This emerging area of health care has the ability to enhance self-management of chronic disease by making health care more accessible, affordable, and available to the public. However, evidence-based information on the use of BCTs in eHealth interventions is limited, and particularly so, for individuals living with CVD. OBJECTIVE: The aim of this systematic review was to assess the application of BCTs in eHealth interventions designed to increase physical activity (PA) in CVD populations. METHODS: A total of 7 electronic databases, including EBSCOhost (MEDLINE, PsycINFO, Academic Search Complete, SPORTDiscus with Full Text, and CINAHL Complete), Scopus, and Web of Science (Core Collection) were searched. Two authors independently reviewed references using the software package Covidence (Veritas Health Innovation). The reviewers met to resolve any discrepancies, with a third independent reviewer acting as an arbitrator when required. Following this, data were extracted from the papers that met the inclusion criteria. Bias assessment of the studies was carried out using the Cochrane Collaboration's tool for assessing the risk of bias within Covidence; this was followed by a narrative synthesis. RESULTS: Out of the 987 studies that were identified, 14 were included in the review. An additional 9 studies were added following a hand search of review paper references. The average number of BCTs used across the 23 studies was 7.2 (range 1-19). The top three most frequently used BCTs included information about health consequences (78%, 18/23), goal setting (behavior; 74%, 17/23), and joint third, self-monitoring of behavior and social support (practical) were included in 11 studies (48%, 11/23) each. CONCLUSIONS: This systematic review is the first to investigate the use of BCTs in PA eHealth interventions specifically designed for people with CVD. This research will have clear implications for health care policy and research by outlining the BCTs used in eHealth interventions for chronic illnesses, in particular CVD, thereby providing clear foundations for further research and developments in the area.

Automatic detection, extraction and analysis of unrestrained gait using a wearable sensor system.

Within this paper we demonstrate the effectiveness of a novel body-worn gait monitoring and analysis framework to both accurately and automatically assess gait during `free-living' conditions. Key features of the system include the ability to automatically identify individual steps within specific gait conditions, and the implementation of continuous waveform analysis within an automated system for the generation of temporally normalized data and their statistical comparison across subjects.

Comparison of discrete-point vs. dimensionality-reduction techniques for describing performance-related aspects of maximal vertical jumping.

The aim of this study was to assess and compare the ability of discrete point analysis (DPA), functional principal component analysis (fPCA) and analysis of characterizing phases (ACP) to describe a dependent variable (jump height) using vertical ground reaction force curves captured during the propulsion phase of a countermovement jump. FPCA and ACP are continuous data analysis techniques that reduce the dimensionality of a data set by identifying phases of variation (key phases), which are used to generate subject scores that describe a subject's behavior. A stepwise multiple regression analysis was used to measure the ability to describe jump height of each data analysis technique. Findings indicated that the order of effectiveness (high to low) across the examined techniques was: ACP (99%), fPCA (78%) and DPA (21%). DPA was outperformed by fPCA and ACP because it can inadvertently compare unrelated features, does not analyze the whole data set and cannot examine important features that occur solely as a phase. ACP outperformed fPCA because it utilizes information within the combined magnitude-time domain, and identifies and examines key phases separately without the deleterious interaction of other key phases.

The variance needed to accurately describe jump height from vertical ground reaction force data.

In functional principal component analysis (fPCA) a threshold is chosen to define the number of retained principal components, which corresponds to the amount of preserved information. A variety of thresholds have been used in previous studies and the chosen threshold is often not evaluated. The aim of this study is to identify the optimal threshold that preserves the information needed to describe a jump height accurately utilizing vertical ground reaction force (vGRF) curves. To find an optimal threshold, a neural network was used to predict jump height from vGRF curve measures generated using different fPCA thresholds. The findings indicate that a threshold from 99% to 99.9% (6-11 principal components) is optimal for describing jump height, as these thresholds generated significantly lower jump height prediction errors than other thresholds.

Clustering vertical ground reaction force curves produced during countermovement jumps.

The aim of this study is to assess and compare the performance of commonly used hierarchical, partitional (k-means) and Gaussian model-based (Expectation-Maximization algorithm) clustering techniques to appropriately identify subgroup patterns within vertical ground reaction force data, using a continuous waveform analysis. In addition, we also compared the performance across each technique using normalized and non-normalization input scores. Both generated and real data (one hundred and twenty two vertical jumps) were analyzed. The performance of each cluster technique was measured by assessing the ability to explain variances in jump height using a stepwise regression analysis. Only k-means (normalized scores; 82%) and hierarchical clustering (normalized scores; 85%) were able to extend the ability to describe variances in jump height beyond that achieved using the group analysis (i.e. one cluster; 78%). Further, our findings strongly indicate the need to normalize the input data (similarity measure) when clustering. In contrast to the group analysis, the subgroup analysis was able to identify cluster specific phases of variance, which improved the ability to explain variances in jump height, due to the identification of cluster specific predictor variables. Our findings therefore highlight the benefit of performing a subgroup analysis and may explain, at least in part, the contrasting findings between previous studies that used a single group level of analysis.

Analysis of characterizing phases on waveform: an application to vertical jumps.

The aim of this study is to propose a novel data analysis approach, an analysis of characterizing phases (ACP), that detects and examines phases of variance within a sample of curves utilizing the time, magnitude, and magnitude-time domains; and to compare the findings of ACP to discrete point analysis in identifying performance-related factors in vertical jumps. Twenty-five vertical jumps were analyzed. Discrete point analysis identified the initial-to-maximum rate of force development (P=.006) and the time from initial-to-maximum force (P=.047) as performance-related factors. However, due to intersubject variability in the shape of the force curves (ie, non-, uni- and bimodal nature), these variables were judged to be functionally erroneous. In contrast, ACP identified the ability to apply forces for longer (P<.038), generate higher forces (P<.027), and produce a greater rate of force development (P<.003) as performance-related factors. Analysis of characterizing phases showed advantages over discrete point analysis in identifying performance-related factors because it (i) analyses only related phases, (ii) analyses the whole data set, (iii) can identify performance-related factors that occur solely as a phase, (iv) identifies the specific phase over which differences occur, and (v) analyses the time, magnitude and combined magnitude-time domains.

Portable integrated microfluidic analytical platform for the monitoring and detection of nitrite.

A wireless, portable, fully-integrated microfluidic analytical platform has been developed and applied to the monitoring and determination of nitrite anions in water, using the Griess method. The colour intensity of the Griess reagent nitrite complex is detected using a low cost Paired Emitter Detector Diode, while on-chip fluid manipulation is performed using a biomimetic photoresponsive ionogel microvalve, controlled by a white light LED. The microfluidic analytical platform exhibited very low limits of detection (34.0±0.1 µg L(-1) of NO2(-)). Results obtained with split freshwater samples showed good agreement between the microfluidic chip platform and a conventional UV-vis spectrophotometer (R(2)=0.98, RSD=1.93% and R(2)=0.99, RSD=1.57%, respectively). The small size, low weight, and low cost of the proposed microfluidic platform coupled with integrated wireless communications capabilities make it ideal for in situ environmental monitoring. The prototype device allows instrument operational parameters to be controlled and analytical data to be downloaded from remote locations. To our knowledge, this is the first demonstration of a fully functional microfluidic platform with integrated photo-based valving and photo-detection.