Home monitoring of respiration in Covid-19 patients using smartphone technology
Funder: Science Foundation Ireland/Enterprise Ireland/IDA Ireland joint Covid-19 Rapid Response Fund
Biomedical engineers at UCD, Professor Madeleine Lowery and Dr Emer Doheny gained SFI-EI-IDA Rapid Response funding to develop a system to remotely monitor respiratory rate in patients with Covid-19 using smartphone microphones and predictive models of hospital admissions.
Once developed, the new system will integrate with the existing Covid19-App – designed by patientMpower – to enable better management of self-isolating patients and prioritisation of hospital admissions.
Remote monitoring of Covid-19 patients who are self-isolating is critical for management of patient symptoms, early identification of patients whose symptoms are deteriorating and prioritisation of hospital admissions.
From the outset, the HSE’s Covid-19App enabled clinicians to remotely monitor oxygen saturation, pulse rate and body temperature, and patient-reported breathlessness via a dashboard. However, it did not monitor respiratory rate.
Respiratory distress is a characteristic symptom of Covid‐19, with most patients admitted to intensive care unable to breathe spontaneously. The most common reason for requiring intensive care has been respiratory support, with two-thirds experiencing acute respiratory distress syndrome.
Due to the large numbers of self-isolating patients, the lack of a reliable, quantitative method to remotely monitor respiration rate is a significant issue.
Accurate remote monitoring of respiration is difficult, with self-reported respiration rate or breathlessness being current practice for phone assessment. However, these methods to self-report breathlessness are subjective, difficult to standardise across patients or time, and unlikely to capture subtle day-to-day changes.
What will the research project do?
Using the microphone of their smartphone, patients will record their breathing using a clearly defined protocol. Algorithms to extract respiration rate and related metrics from these audio recordings will be developed, providing additional vital signs monitoring to clinical data currently being monitored remotely.
Additionally, predictive models will be developed to support early identification of at-risk patients who require hospitalisation. To develop a robust risk assessment algorithm, the extracted respiratory features will be combined with other data collected by the Covid-19 Patient Management App using feature selection and supervised machine learning, along with appropriate validation methods.
The solution will allow clinicians to remotely monitor respiration rate in Covid-19 patients in a reliable, scalable and affordable manner, requiring only a smartphone. The system will integrate with an existing platform currently in use by hundreds of patients in Ireland.
The solution will also provide a tool to predict hospitalisation in large cohorts of patients, improving patient care and enabling more effective allocation of hospital resources and reducing hospital self-isolating, and thereby hospital visits.
While the proposed study is motivated by the current Covid-19 pandemic, measurement of respiration rate using smartphones and early identification of patients at risk of respiratory distress has widespread value for remote monitoring of patients with chronic respiratory disease generally.
Eamonn Costello, CEO, patientMpower Ltd
Colin Edwards, CSO, patientMpower Ltd
Dr Silke Ryan, Consultant in Respiratory Medicine, St Vincent’s University Hospital (SVUH)
Dr Cormac McCarthy, Consultant in Respiratory Medicine, SVUH
Professor Patrick Mallon, Consultant in Infectious Diseases, SVUH
Dr Eoin Feeney, Consultant in Infectious Diseases, SVUH
Dr Aoife Cotter, Consultant in Infectious Diseases, Mater Misericordiae University Hospital (MMUH)
Dr Tara McGinty, Consultant in Infectious Diseases, MMUH