Windowing is an established technique employed within dense sensing environments to extract relevant features from sensor data streams. Among the established approaches of Explicit, Time-based and Sensor-Event based windowing, Dynamic windowing approaches are beginning to emerge. These dynamic approaches claim to address the inherent shortcomings of the aforementioned established approaches by determining the appropriate window length for live sensor data streams in real-time, thereby offering the potential to optimize and increase the recognition of these sensor represented activities. Beyond these potential benefits, dynamic approaches can also support anomaly detection by actively uncovering new, unknown window patterns within a trained model. This paper presents findings from a study which utilizes data from a single source dataset, towards benchmarking and comparing more traditional windowing approaches against a dynamic windowing approach. The experiments conducted on a real-world smart home dataset suggest Time-based windowing is the best approach. Through evaluation of results, Dynamic windowing approaches may benefit from carefully annotated datasets.
Background: In this study we propose the development of a new algorithm for selecting optimal recording sites for limited lead body surface potential mapping. The proposed algorithm differs from previously reported methods in that it is based upon a simple and intuitive data driven technique that does not make any presumptions about deterministic characteristics of the data. It uses a forward selection based search technique to find the best combination of electrocardiographic leads. Methods: The study was conducted using a dataset consisting of body surface potential maps (BSPM) recorded from 116 subjects which included 59 normals and 57 subjects exhibiting evidence of old Myocardial Infarction (MI). The performance of the algorithm was evaluated using spatial RMS voltage error and correlation coefficient to compare original and reconstructed map frames. Results: In all, three configurations of the algorithm were evaluated and it was concluded that there was little difference in the performance of the various configurations. In addition to observing the performance of the selection algorithm, several lead subsets of 32 electrodes as chosen by the various configurations of the algorithm were evaluated. The rationale for choosing this number of recording sites was to allow comparison with a previous study that used a different algorithm, where 32 leads were deemed to provide an acceptable level of reconstruction performance. Conclusion: It was observed that although the lead configurations suggested in this study were not identical to that suggested in the previous work, the systems did bear similar characteristics in that recording sites were chosen with greatest density in the precordial region.
Background: People with autism spectrum disorder (ASD) frequently experience high levels of anxiety. Despite this, many clinical settings do not provide specialist ASD mental health services, and demand for professional support frequently outstrips supply. Across many sectors of health, investigators have explored digital health solutions to mitigate demand and extend the reach of professional practice beyond traditional clinical settings. Objective: This critical appraisal and pilot feasibility study examines heart rate variability (HRV) biofeedback as an approach to help young people with ASD to manage anxiety symptoms outside of formal settings. The aim is to explore the use of portable biofeedback devices to manage anxiety, while also highlighting the risks and benefits of this approach with this population. Methods: We assessed the feasibility of using home-based HRV biofeedback for self-management of anxiety in young people with ASD. We adopted coproduction, involving people with ASD, to facilitate development of the study design. Next, a separate pilot with 20 participants with ASD (n=16, 80% male participants and n=4, 20% female participants, aged 13-24 years; IQ>70) assessed adoption and acceptability of HRV biofeedback devices for home use over a 12-week period. Data were collected from both carers and participants through questionnaires and interviews; participants also provided single-lead electrocardiogram recordings as well as daily reports through smartphone on adoption and use of their device. Results: Pre-post participant questionnaires indicated a significant reduction in anxiety in children (t6=2.55; P=.04; Cohen d=0.99) as well as adults (t7=3.95; P=.006; Cohen d=0.54). Participant age was significantly negatively correlated with all HRV variables at baseline, namely high-frequency heart rate variability (HF-HRV: P=.02), the root mean square of successive differences in normal heartbeat contractions (RMSSD: P=.02) and the variability of normal-to-normal interbeat intervals (SDNN: P=.04). At follow-up, only SDNN was significantly negatively correlated with age (P=.05). Levels of ASD symptoms were positively correlated with heart rate both before (P=.04) and after the intervention (P=.01). The majority (311/474, 65.6%) of reports from participants indicated that the devices helped when used. Difficulties with the use of some devices and problems with home testing of HRV were noted. These initial findings are discussed within the context of the strengths and challenges of remotely delivering a biofeedback intervention for people with ASD. Conclusions: HRV biofeedback devices have shown promise in this pilot study. There is now a need for larger evaluation of biofeedback to determine which delivery methods achieve the greatest effect for people with ASD. Trial Registration: ClinicalTrials.gov NCT04955093; https://clinicaltrials.gov/ct2/show/NCT04955093
Data driven approaches for human activity recognition learn from pre-existent large-scale datasets to generate a classification algorithm that can recognize target activities. Typically, several activities are represented within such datasets, characterized by multiple features that are computed from sensor devices. Often, some features are found to be more relevant to particular activities, which can lead to the classification algorithm providing less accuracy in detecting the activity where such features are not so relevant. This work presents an experimentation for human activity recognition with features derived from the acceleration data of a wearable device. Specifically, this work analyzes which features are most relevant for each activity and furthermore investigates which classifier provides the best accuracy with those features. The results obtained indicate that the best classifier is the k-nearest neighbor and furthermore, confirms that there do exist redundant features that generally introduce noise into the classification, leading to decreased accuracy.
Introduction Speech and language therapists (SLTs) worldwide report challenges with providing recommended, evidence-based intervention intensity for children with speech sound disorder (SSD). Challenges such as service constraints and/or family contexts impact on access to optimal therapy intensity. Existing research indicates that empowering and training parents to deliver intervention at home, alongside SLT support, offers one possible solution to increasing the intensity of intervention children with SSD receive. Digital health could increase accessibility to intensive home practice and help sustain engagement with therapy activities. Further exploration is needed around what makes parent-implemented interventions for children with SSD effective, for who and in which situations. This paper outlines the protocol for a realist review which aims to explore the active ingredients and contextual factors of effective digital parent-led interventions. Methods and analysis A realist review will explore the research question, following six stages. The scope of the review will be determined, and initial programme theories will be developed about what works in digital parent-implemented interventions for SSD, for whom, how, why and in what circumstances. Relevant secondary data, identified through a formal search strategy, will be selected, appraised, analysed and synthesised using realist principles to test and further refine the initial programme theories. This process will develop refined underpinning explanatory theories which capture the interaction between contexts, mechanisms and outcomes of the intervention. An expert steering group will provide insight to inform explanatory theories, searches, and dissemination. Ethics and dissemination Ethical approval is not required for this review. The refined programme theories from the review will inform the next stages of a wider study. A subsequent realist evaluation will test and further refine theories with key stakeholders. Following this, the underpinning programme theory will be used to coproduce a digital tool, to support parents to deliver home intervention alongside SLT support.