Presentation of long COVID and associated risk factors in a mobile health study (preprint)

Background The Covid Collab study was a citizen science mobile health research project set up in June 2020 to monitor COVID-19 symptoms and mental health through questionnaire self-reports and passive wearable device data. Methods Using mobile health data, we consider whether a participant is suffering from long COVID in two ways. Firstly, by whether the participant has a persistent change in a physiological signal commencing at a diagnosis of COVID-19 that last for at least twelve weeks. Secondly, by whether a participant has self-reported persistent symptoms for at least twelve weeks. We assess sociodemographic and wearable-based risk factors for the development of long COVID according to the above two categorisations. Findings Persistent changes to physiological signals measured by com- mercial fitness wearables, including heart rate, sleep, and activity, are visible following a COVID-19 infection and may help differentiate people who develop long COVID. Anxiety and depression are significantly and persistently affected at a group level following a COVID-19 infection. We found the level of activity undertaken in the year prior to illness was protective against long COVID and that symptoms of depression before and during the acute illness may be a risk factor. Interpretation Mobile health and wearable devices may prove to be a useful resource for tracking recovery and presence of long-term sequelae to COVID-19. Mental wellbeing is significantly negatively effected on average for an extended period of time following a COVID-19 infection.

Fitbeat: COVID-19 Estimation based on Wristband Heart Rate (preprint)

This study investigates the potential of deep learning methods to identify individuals with suspected COVID-19 infection using remotely collected heart-rate data. The study utilises data from the ongoing EU IMI RADAR-CNS research project that is investigating the feasibility of wearable devices and smart phones to monitor individuals with multiple sclerosis (MS), depression or epilepsy. Aspart of the project protocol, heart-rate data was collected from participants using a Fitbit wristband. The presence of COVID-19 in the cohort in this work was either confirmed through a positive swab test, or inferred through the self-reporting of a combination of symptoms including fever, respiratory symptoms, loss of smell or taste, tiredness and gastrointestinal symptoms. Experimental results indicate that our proposed contrastive convolutional auto-encoder (contrastive CAE), i. e., a combined architecture of an auto-encoder and contrastive loss, outperforms a conventional convolutional neural network (CNN), as well as a convolutional auto-encoder (CAE) without using contrastive loss. Our final contrastive CAE achieves 95.3% unweighted average recall, 86.4% precision, anF1 measure of 88.2%, a sensitivity of 100% and a specificity of 90.6% on a testset of 19 participants with MS who reported symptoms of COVID-19. Each of these participants was paired with a participant with MS with no COVID-19 symptoms.

Using smartphones and wearable devices to monitor behavioural changes during COVID-19 (preprint)

We aimed to explore the utility of the recently developed open-source mobile health platform RADAR-base as a toolbox to rapidly test the effect and response to NPIs aimed at limiting the spread of COVID-19. We analysed data extracted from smartphone and wearable devices and managed by the RADAR-base from 1062 participants recruited in Italy, Spain, Denmark, the UK, and the Netherlands. We derived nine features on a daily basis including time spent at home, maximum distance travelled from home, maximum number of Bluetooth-enabled nearby devices (as a proxy for physical distancing), step count, average heart rate, sleep duration, bedtime, phone unlock duration, and social app use duration. We performed Kruskal-Wallis tests followed by post-hoc Dunns tests to assess differences in these features among baseline, pre-, and during-lockdown periods. We also studied behavioural differences by age, gender, body mass index (BMI), and educational background. We were able to quantify expected changes in time spent at home, distance travelled, and the number of nearby Bluetooth-enabled devices between pre- and during-lockdown periods. We saw reduced sociality as measured through mobility features, and increased virtual sociality through phone usage. People were more active on their phones, spending more time using social media apps, particularly around major news events. Furthermore, participants had lower heart rate, went to bed later, and slept more. We also found that young people had longer homestay than older people during lockdown and fewer daily steps. Although there was no significant difference between the high and low BMI groups in time spent at home, the low BMI group walked more. RADAR-base can be used to rapidly quantify and provide a holistic view of behavioural changes in response to public health interventions as a result of infectious outbreaks such as COVID-19.