There have been sustained efforts toward using naturalistic methods in developmental science to measure infant behaviors in the real world from an egocentric perspective because statistical regularities in the environment can shape and be shaped by the developing infant. However, there is no user-friendly and unobtrusive technology to densely and reliably sample life in the wild. To address this gap, we present the design, implementation and validation of the EgoActive platform, which addresses limitations of existing wearable technologies for developmental research. EgoActive records the active infants’ egocentric perspective of the world via a miniature wireless head-mounted camera concurrently with their physiological responses to this input via a lightweight, wireless ECG/acceleration sensor. We also provide software tools to facilitate data analyses. Our validation studies showed that the cameras and body sensors performed well. Families also reported that the platform was comfortable, easy to use and operate, and did not interfere with daily activities. The synchronized multimodal data from the EgoActive platform can help tease apart complex processes that are important for child development to further our understanding of areas ranging from executive function to emotion processing and social learning.

Infant electrocardiograms (ECGs) and heart rates (HRs) are very useful biosignals for psychological research and clinical work, but can be hard to analyse properly, particularly longform (≥5 min) recordings taken in naturalistic environments. Infant HRs are typically much faster than adult HRs, and so some of the underlying frequency assumptions made about adult ECGs may not hold for infants. However, the bulk of publicly available ECG approaches focus on adult data. Here, existing open source ECG approaches are tested on infant datasets. The best-performing open source method is then modified to maximise its performance on infant data (e.g., including a 15 Hz high-pass filter, adding local peak correction). The HR signal is then subsequently analysed, developing an approach for cleaning data with separate sets of parameters for the analysis of cleaner and noisier HRs. A Signal Quality Index (SQI) for HR is also developed, providing insights into where a signal is recoverable and where it is not, allowing for more confidence in the analysis performed on naturalistic recordings. The tools developed and reported in this paper provide a base for the future analysis of infant ECGs and related biophysical characteristics. Of particular importance, the proposed solutions outlined here can be efficiently applied to real-world, large datasets.

Introduction Adherence is often cited as an important metric to demonstrate sustained engagement of an individual or population with a health technology, but its definition is often ill-defined. Any adherence definition made for digital health interventions must be clearly defined to ensure a consistent approach to measuring sustained use as an indicator of impact. Methods This study followed mathematically-defined definitions of distinct aspects of adherence: initial adoption, consistency, duration, and dropout. These were then applied to a digital physiotherapy dataset of older adults (N=56). Participants were assigned 3 sessions a week of exergames (exercise-based videogames) for 12 weeks using MIRA rehab software platform. Results The following adherence characteristics emerged: an initial dropout of 3% (completed ≤3 sessions), 20% of participants achieving the desired consistency (≥3 sessions a week for 12 weeks), 39% of participants passing a duration threshold (completing ≥20 minutes a week for 12 weeks), and an average dropout at 72.3% (when judged by percentage of sessions completed at dropout). Conclusion The approach used for measuring and reporting adherence metrics allows readers to draw clear conclusions about the different aspects of engagement that users displayed with the digital health programme. This type of reporting is recommended for all future digital health studies reporting adherence measures to ensure a consistent approach to reporting and comparing digital health interventions and their impact.

Infant electrocardiograms (ECG) and heart rates (HR) are very useful biosignals for psychological research and clinical work, but can be hard to analyze properly, particularly long form (≥5 minutes) recordings taken in naturalistic environments. Infant HRs are typically much faster than adult HRs, and so some of the underlying frequency assumptions made about adult ECGs may not hold for infants. However, the bulk of publicly available ECG approaches focus on adult data. Here, existing open-source ECG approaches are tested on infant datasets. The best performing open-source method is then modified to maximize its performance on infant data (e.g., including a 15Hz high pass filter, adding local peak correction). The HR signal is then subsequently analyzed, developing an approach for cleaning data with separate sets of parameters for the analysis of cleaner and noisier HR. A Signal Quality Index (SQI) for HR is also developed, providing insight into where a signal is recoverable and where it is not, allowing for more confidence in analysis performed on naturalistic recordings. The tools developed and reported in this paper provide a base for future analysis of infant ECG and related biophysical characteristics. Of particular importance, the proposed solutions outlined here can be efficiently applied to real-world large datasets.

Acceleration methods in fMRI aim to reconstruct high fidelity images from undersampled k-space, allowing fMRI datasets to achieve higher temporal resolution, reduced physiological noise aliasing, and increased statistical degrees of freedom. While low levels of acceleration are typically part of standard fMRI protocols through parallel imaging, there exists the potential for approaches that allow much greater acceleration. One such existing approach is k-t FASTER, which exploits the inherent low-rank nature of fMRI. In this paper, we present a reformulated version of k-t FASTER which includes additional L2 constraints within a low-rank framework. We evaluated the effect of three different constraints against existing low-rank approaches to fMRI reconstruction: Tikhonov constraints, low-resolution priors, and temporal subspace smoothness. The different approaches are separately tested for robustness to undersampling and thermal noise levels, in both retrospectively and prospectively-undersampled finger-tapping task fMRI data. Reconstruction quality is evaluated by accurate reconstruction of low-rank subspaces and activation maps. The use of L2 constraints was found to achieve consistently improved results, producing high fidelity reconstructions of statistical parameter maps at higher acceleration factors and lower SNR values than existing methods, but at a cost of longer computation time. In particular, the Tikhonov constraint proved very robust across all tested datasets, and the temporal subspace smoothness constraint provided the best reconstruction scores in the prospectively-undersampled dataset. These results demonstrate that regularized low-rank reconstruction of fMRI data can recover functional information at high acceleration factors without the use of any model-based spatial constraints.

Background: The long-term effects of a COVID-19 diagnosis on respiratory outcomes by vaccination status, severity of COVID-19, and pre-existing conditions remain unclear. This study investigated the association between COVID-19 and respiratory conditions, considering these factors. Methods: OpenSAFELY-TPP was used to access linked data from 18 million people registered with GPs in England. Three cohorts were defined: pre-vaccination (wild-type/Alpha era), vaccinated and unvaccinated (Delta era). Adjusted hazard ratios (aHRs) compared the incidence of respiratory conditions after COVID-19 with that before or without COVID-19 for pneumonia, asthma, chronic obstructive pulmonary disease (COPD), and interstitial lung disease (ILD). Findings: Incidence of all outcomes was elevated during weeks 1-4 after COVID-19, compared with before or without COVID-19. Although attenuated beyond four weeks, incidence remained elevated for up to 5·5 years after COVID-19 for some outcomes, particularly following hospitalised COVID-19, e.g., COPD: pre-vaccination aHR 1·49[95% CI, 1·24-1·80], vaccinated 1·66[1·25-2·20], unvaccinated 2·15[1·41-3·28]. In most cases, aHRs were lower in vaccinated than unvaccinated people. Absolute incidence of pneumonia and ILD was higher among those with than without pre-existing respiratory conditions. Interpretation: Incidence of respiratory disease remains elevated particularly in the first month, irrespective of vaccination status and pre-existing conditions and for up to five years following hospitalised COVID-19. These findings have important implications for post-COVID clinical care, highlighting the need for long-term respiratory follow-up and healthcare service planning. Funding: UKRI Medical Research Council and UK National Institute for Health and Care Research.