BACKGROUND
The traditional measurement of these vitals via physical examination has become challenging and the recent pandemic has accelerated trends towards telehealth and remote monitoring. Instead of going to the physician for checking heart rate, SPO2, blood pressure, body temperature and respiration rate it would be excellent if it could be measured at home. Vital signs monitors, also known as physiological parameter monitors, are electronic devices that measure and display biological information about patients under constant monitoring.
OBJECTIVE
The purpose of this study was to validate the accuracy of the Pulse Oxygen Saturation, Heart Rate, Blood Pressure and Respiratory Rate of Docsun Telehealth by comparing it with approved medical devices.
METHODS
This is a non-invasive self-check system-based study which was conducted to validate the detection of vital signs (Pulse Oxygen Saturation, Heart Rate, Blood Pressure and Respiratory Rate) raised by the Docsun Telehealth Portal. Data for this study was collected at one collaborating site approved by Docsun. The data of patients visiting the site was collected based on the study criteria after obtaining their informed consent. This study aims to collect basic vital sign information on the current health status for generating a diagnostic report by the AI tool. The input for the software processing is a facial screening without any accessories on the face, directly scanning through the software application portal. The subject’s facial features are detected and screened for the extraction of necessary reading. The study setup was designed such that the subject was always at a distance of 2m from the study staff. The brochures and subject’s signature log was placed near the study screening space. Basic vital signs like Pulse Oxygen Saturation, Heart Rate, Blood Pressure and Respiratory Rate were measured using pulse oximeters and digital blood pressure monitors. The readings were noted in a case report form. A webcam was installed at a convenient distance from the subject for the facial screening. Then the subject was asked to look into the webcam for software screening. The subject asked not to make any sudden movements but just normal breathing for a time period of 60 seconds. The software would then provide a health report and its upload time was noted for study data collection purposes. The study area of the subject was then be sanitized using a spray and fumigation machine before the entry of the next subject.
RESULTS
For the HR, SPO2, BP and RR measurement validation, the main outcome was the mean of the absolute difference between the respective investigational devices and the reference values and the percent absolute difference between the respective investigational devices and the reference values.
HR measurements were considered accurate if the mean absolute difference was within either ±10 % or ± 5 bpm, depending on which of the two was greater. RR measurements were considered accurate if the mean absolute difference was within either ±10 % or ± 3 bpm. If the HR was within ±10% of the reference standard or 5 beats/min, it was considered to be acceptable for clinical purposes. The average absolute difference between the Docsun Telehealth Portal and the reference values was 1.41 bpm. The mean absolute percentage difference was 1.69%. The Docsun Telehealth Portal therefore met the predefined accuracy cutoff for HR measurements.
If the RR was within ±10% of the reference standard or 3 breaths/min, it was considered to be acceptable for clinical purposes. The average absolute difference between the Docsun Telehealth Portal was 0.86 breaths/minute. The absolute percentage difference was 4.72%. The Docsun Telehealth Portal therefore met the predefined accuracy cutoff for RR measurements.
SPO2 levels were considered acceptable if the average absolute difference between the Docsun Telehealth Portal and the reference values was ±3%. The mean absolute percentage difference was 0.59%. The Docsun Telehealth Portal therefore met the predefined accuracy cutoff for SPO2 measurements.
The Docsun Telehealth Portal predicted systolic BP with an accuracy of 94.81% and diastolic BP with an accuracy of 95.71%. The average prediction bias±error SDs were 0.39±7.30 mm Hg for systolic BP and −0.20±6.00 mm Hg for diastolic BP. These SDs represent information gains of 25.5%, 12.0%, and 21.8%, respectively. Our findings corresponded to average intraclass correlations of 0.60, 0.37, and 0.56 and average Pearson correlations of 0.67 and 0.47, for systolic and diastolic, respectively.
CONCLUSIONS
The results of the study shows that the accuracy of the Heart Rate, Blood Pressure, SPO2 and Breathing Rate values raised by the Docsun Telehealth Portal compared against the clinically approved medical device proceed to be accurate by meeting the predefined accuracy guidelines.