There is evidence that patients with COVID-19 can be managed at home, however over the phone or video consultation are difficult to scale. We propose to use a mobile application to collect data and present data to clinians to improve the management of COVID-19 patients at home and to predict deterioration of people with test positive COVID-19 infection enabling timely appropriate triage of affected patients while providing reassurance to others.
Our team’s track record of using mobile devices to monitor patients with Chronic Obstructive Pulmonary Disease (COPD) since 2014 prepares us to conduct this project: First, we have the infrastructure and logistical experience to conduct health research using mobile devices (e.g. setting up/managing servers at UHN, preparing and sending out devices to patients, etc.). Second, we have mobile and smartwatch applications that provide near continuous monitoring of heart rate, motion, steps and physical activity that upload the data to a secure cloud server. Third, we have developed methods to infer respiratory rate and coughing. Fourth, we have procedures for managing highly sensitive personal health information and developed methods, such as audio muffling, to address privacy concerns. Finally, we have developed a dashboard for real time visualization of patient data and alerting clinicians of any abnormalities.
The technology combines novel filters and a convolutional neural network model to automatically learn what kind of data will make an algorithm produce inaccurate results. While the idea of rejecting certain data is not new, existing filters are manually-developed and rely on assumptions about what causes the algorithm to be unreliable.
The approach applies machine learning methods to automatically learn when sensor processing algorithms will be reliable or unreliable and discards unreliable data, making the overall system more accurate.
In-the-wild continuous sensing on mobile devices has the potential to revolutionize fields such as personalized health care. However, a key problem with current methods is the diverse nature and noise associated with incoming sensor data. Running sensor data processing algorithms on this diverse data can lead to unexpected and poor results because it is difficult for algorithms to anticipate the variety of data that can occur. This tunable system is able to achieve error rates significantly lower than existing approaches, especially for:
- At‑home monitoring
- Wearable tech
- Clinical studies
Patent application has been filed and a proof of concept has been created and tested in a clinical setting with patients who have COPD by implementing a respiratory rate monitor on a smartwatch. Existing smartwatch-based respiratory rate monitors rely on the assumption that motion will cause the respiratory rate monitor to be unreliable. This filter automatically learns when respiratory rate detection from a smartwatch will be reliable and yields significantly better results than current methods.