Prediction of low pulse oxygen saturation in COVID-19 using remote monitoring post hospital discharge.

BACKGROUND: Monitoring systems have been developed during the COVID-19 pandemic enabling clinicians to remotely monitor physiological measures including pulse oxygen saturation (SpO2), heart rate (HR), and breathlessness in patients after discharge from hospital. These data may be leveraged to understand how symptoms vary over time in COVID-19 patients. There is also potential to use remote monitoring systems to predict clinical deterioration allowing early identification of patients in need of intervention. METHODS: A remote monitoring system was used to monitor 209 patients diagnosed with COVID-19 in the period following hospital discharge. This system consisted of a patient-facing app paired with a Bluetooth-enabled pulse oximeter (measuring SpO2 and HR) linked to a secure portal where data were available for clinical review. Breathlessness score was entered manually to the app. Clinical teams were alerted automatically when SpO2 < 94 %. In this study, data recorded during the initial ten days of monitoring were retrospectively examined, and a random forest model was developed to predict SpO2 < 94 % on a given day using SpO2 and HR data from the two previous days and day of discharge. RESULTS: Over the 10-day monitoring period, mean SpO2 and HR increased significantly, while breathlessness decreased. The coefficient of variation in SpO2, HR and breathlessness also decreased over the monitoring period. The model predicted SpO2 alerts (SpO2 < 94 %) with a mean cross-validated. sensitivity of 66 ± 18.57 %, specificity of 88.31 ± 10.97 % and area under the receiver operating characteristic of 0.80 ± 0.11. Patient age and sex were not significantly associated with the occurrence of asymptomatic SpO2 alerts. CONCLUSION: Results indicate that SpO2 alerts (SpO2 < 94 %) on a given day can be predicted using SpO2 and heart rate data captured on the two preceding days via remote monitoring. The methods presented may help early identification of patients with COVID-19 at risk of clinical deterioration using remote monitoring.

Estimation of respiratory rate and exhale duration using audio signals recorded by smartphone microphones

Monitoring respiration using mobile technology has potential to contribute to the clinical management of patients with infectious diseases or chronic respiratory conditions in the home. In this study, a new method to estimate respiratory rate and exhale duration from audio data recorded using smartphone microphones was developed. The method first determines the fundamental frequency of the audio signal, which guides an adaptive thresholding method to detect individual exhales. Exhale boundary times were refined using adaptive physiological thresholds. To control for environmental noise in remote audio recordings, a method to classify audio signals as acceptable or unacceptable for accurate respiration monitoring was developed. Estimated respiratory rates and audio exhale durations were validated against respiratory inductance plethysmography (RIP) in 27 healthy participants. A further 217 audio recordings were collected remotely by 210 healthy and COVID-19 participants, with results compared with researcher annotations. Compared to RIP in the laboratory, respiratory rate was estimated with a mean absolute error (MAE) of 0.2 ± 0.27 bpm, and within 1 bpm for 96 % of recordings (r = 0.99). Compared to researcher annotations for remote recordings, respiratory rate was estimated with a MAE of 0.79 ± 2.44 bpm, and within 1 bpm for 87.5 % of recordings (r = 0.92), while audio exhale duration was estimated with a MAE of 0.21 ± 0.23 s. Audio signal quality was classified with an area under the receiver operating characteristic of 0.81 ±. This method offers the potential for accurate remote monitoring of respiratory rate and breathing patterns in individuals with COVID-19 and other chronic respiratory conditions.