Background The detection of early changes in vital signs (VSs) enables timely intervention; however, the measurement of VSs requires hands-on technical expertise and is often time-consuming. The contactless measurement of VSs is beneficial to prevent infection, such as during the COVID-19 pandemic. Lifelight is a novel software being developed to measure VSs by remote photoplethysmography based on video captures of the face via the integral camera on mobile phones and tablets. We report two early studies in the development of Lifelight. Objective The objective of the Vital Sign Comparison Between Lifelight and Standard of Care: Development (VISION-D) study (NCT04763746) was to measure respiratory rate (RR), pulse rate (PR), and blood pressure (BP) simultaneously by using the current standard of care manual methods and the Lifelight software to iteratively refine the software algorithms. The objective of the Vital Sign Comparison Between Lifelight and Standard of Care: Validation (VISION-V) study (NCT03998098) was to validate the use of Lifelight software to accurately measure VSs. Methods BP, PR, and RR were measured simultaneously using Lifelight, a sphygmomanometer (BP and PR), and the manual counting of RR. Accuracy performance targets for each VS were defined from a systematic literature review of the performance of state-of-the-art VSs technologies. Results The VISION-D data set (17,233 measurements from 8585 participants) met the accuracy targets for RR (mean error 0.3, SD 3.6 vs target mean error 2.3, SD 5.0; n=7462), PR (mean error 0.3, SD 4.0 vs mean error 2.2, SD 9.2; n=10,214), and diastolic BP (mean error −0.4, SD 8.5 vs mean error 5.5, SD 8.9; n=8951); for systolic BP, the mean error target was met but not the SD (mean error 3.5, SD 16.8 vs mean error 6.7, SD 15.3; n=9233). Fitzpatrick skin type did not affect accuracy. The VISION-V data set (679 measurements from 127 participants) met all the standards: mean error −0.1, SD 3.4 for RR; mean error 1.4, SD 3.8 for PR; mean error 2.8, SD 14.5 for systolic BP; and mean error −0.3, SD 7.0 for diastolic BP. Conclusions At this early stage in development, Lifelight demonstrates sufficient accuracy in the measurement of VSs to support certification for a Level 1 Conformité Européenne mark. As the use of Lifelight does not require specific training or equipment, the software is potentially useful for the contactless measurement of VSs by nonclinical staff in residential and home care settings. Work is continuing to enhance data collection and processing to achieve the robustness and accuracy required for routine clinical use. International Registered Report Identifier (IRRID) RR2-10.2196/14326

Background Remote photoplethysmography (rPPG) can record vital signs (VSs) by detecting subtle changes in the light reflected from the skin. Lifelight (Xim Ltd) is a novel software being developed as a medical device for the contactless measurement of VSs using rPPG via integral cameras on smart devices. Research to date has focused on extracting the pulsatile VS from the raw signal, which can be influenced by factors such as ambient light, skin thickness, facial movements, and skin tone. Objective This preliminary proof-of-concept study outlines a dynamic approach to rPPG signal processing wherein green channel signals from the most relevant areas of the face (the midface, comprising the cheeks, nose, and top of the lip) are optimized for each subject using tiling and aggregation (T&A) algorithms. Methods High-resolution 60-second videos were recorded during the VISION-MD study. The midface was divided into 62 tiles of 20×20 pixels, and the signals from multiple tiles were evaluated using bespoke algorithms through weighting according to signal-to-noise ratio in the frequency domain (SNR-F) score or segmentation. Midface signals before and after T&A were categorized by a trained observer blinded to the data processing as 0 (high quality, suitable for algorithm training), 1 (suitable for algorithm testing), or 2 (inadequate quality). On secondary analysis, observer categories were compared for signals predicted to improve categories following T&A based on the SNR-F score. Observer ratings and SNR-F scores were also compared before and after T&A for Fitzpatrick skin tones 5 and 6, wherein rPPG is hampered by light absorption by melanin. Results The analysis used 4310 videos recorded from 1315 participants. Category 2 and 1 signals had lower mean SNR-F scores than category 0 signals. T&A improved the mean SNR-F score using all algorithms. Depending on the algorithm, 18% (763/4212) to 31% (1306/4212) of signals improved by at least one category, with up to 10% (438/4212) improving into category 0, and 67% (2834/4212) to 79% (3337/4212) remaining in the same category. Importantly, 9% (396/4212) to 21% (875/4212) improved from category 2 (not usable) into category 1. All algorithms showed improvements. No more than 3% (137/4212) of signals were assigned to a lower-quality category following T&A. On secondary analysis, 62% of signals (32/52) were recategorized, as predicted from the SNR-F score. T&A improved SNR-F scores in darker skin tones; 41% of signals (151/369) improved from category 2 to 1 and 12% (44/369) from category 1 to 0. Conclusions The T&A approach to dynamic region of interest selection improved signal quality, including in dark skin tones. The method was verified by comparison with a trained observer’s rating. T&A could overcome factors that compromise whole-face rPPG. This method’s performance in estimating VS is currently being assessed. Trial Registration ClinicalTrials.gov NCT04763746; https://clinicaltrials.gov/ct2/show/NCT04763746

BACKGROUND Remote photoplethysmography (rPPG) can record vital signs (VS) by detecting subtle changes in the light reflected from the skin. Lifelight®(Xim Ltd) is a novel software being developed as a medical device for the contactless measurement of VS using rPPG via the integral cameras on smart devices. Research to date has focused on extracting the pulsatile VS signal from the raw signal, which can be influenced by factors such as ambient light, skin thickness, facial movements and skin tone. OBJECTIVE This preliminary proof-of-concept study outlines a dynamic approach to rPPG signal processing in which green channel signals from the most relevant areas of the face (the mid-face, comprising the cheeks, nose and top of the lip) are optimized for each subject using tiling and aggregation (T&A) algorithms. METHODS High-resolution 60 second videos were recorded during the VISION-MD study (Clinicaltrials.gov identifier NCT04763746). The mid-face was divided into 62 tiles of 20 × 20 pixels and the best 30 tiles, based on the signal to noise ratio in the frequency domain (SNR-F), aggregated using five different algorithms. Signals from the mid-face before and after T&A were categorized by a trained observer blinded to the data processing as 0 (high quality, suitable for algorithm training), 1 (suitable for algorithm testing) or 2 (inadequate quality). In a secondary analysis, observer categories were compared for signals predicted to improve category following T&A based on SNR-F score. Observer ratings and SNR-F scores were also compared before and after T&A for Fitzpatrick skin tones 5 and 6, in which rPPG is hampered by light absorption by melanin. RESULTS The analysis used 4310 videos recorded from 1315 participants. Signals in categories 2 and 1 had lower mean SNR-F scores than those in category 0. T&A improved the mean SNR-F score using all algorithms. Nine to 21% improved by at least one category, with up to 10% improving into category 0, and 15–39% remained in the same category,. Importantly, 9–21% improved from category 2 (not usable) into category 1. Improvements were seen with all the algorithms tested. No more than 2% of signals were assigned into a lower-quality category following T&A. In the secondary analysis, 62% of 52 signals were re-categorized by the observer as predicted from SNR-F score. T&A improved SNR-F scores in darker skin tones; 41% of 369 signals improved from category 2 to 1 and 12% from category 1 to 0. CONCLUSIONS The T&A approach to dynamic ROI selection improved signal quality, including in dark skin tones. The method was verified by comparison with a trained observer rating. T&A can reasonably be expected to overcome factors that compromise whole-face rPPG. The performance of this method in estimating VS is currently being assessed.

BACKGROUND Detection of early changes in vital signs (VS) enables timely intervention; however, measurement of VS requires hands-on technical expertise and is often time-consuming. Contactless measurement of VS is beneficial to prevent infection, such as during the COVID-19 pandemic. Lifelight® is a novel software being developed to measure VS by remote photoplethysmography, based on video capture of the face via the integral camera on mobile phones and tablets. We report the observational VISION-D data collection study for algorithm development (NCT04763746) and VISION-V (NCT03998098), a laboratory-based validation study of Lifelight. OBJECTIVE Data collection for algorithm development (VISION D) and software validation (VISION V) METHODS Blood pressure (BP), pulse rate (PR), and respiratory rate (RR) were measured simultaneously using Lifelight, a sphygmomanometer (BP, PR) and manual counting of RR. Accuracy performance targets for each VS were defined from a systematic literature review of the performance of state-of-the-art VS technologies. RESULTS The VISION-D dataset (17,233 measurements from 8585 participants) met the accuracy targets for RR (0.3 ± 3.6 [mean error ± SD] vs target of 2.3 ± 5.0; n=7462), PR (0.3 ± 4.0 vs 2.2 ± 9.2; n=10,214), and diastolic BP (−0.4 ± 8.5 vs 5.5 ± 8.9; n=8951); for systolic BP, mean error target was met but not SD (3.5 ± 16.8 vs 6.7 ± 15.3; n=9233). Fitzpatrick skin type did not affect accuracy. The VISION-V dataset (679 measurements from 127 participants) met all the standards: −0.1 ± 3.4 for RR; 1.4 ± 3.8 for PR; 2.8 ± 14.5 for systolic BP; −0.3 ± 7.0 for diastolic BP. CONCLUSIONS Lifelight demonstrates sufficient accuracy in the measurement of VS, particularly RR and PR, which are important early indicators of clinical deterioration. As use of Lifelight does not require specific training or equipment, the software is potentially useful for the contactless measurement of VS by non-clinical staff in residential and home care settings. INTERNATIONAL REGISTERED REPORT RR2-10.2196/14326