wave-grey
Contact Us

Solution
A Ready-to-Use
Vital Signs Monitoring
Solution for Organizations

remote Photoplethysmography (rPPG )based end-to-end framework Vitalism for remote measurement of vital signs in near real-time using only a face video captured in natural environment using a smartphone camera.

Vitalism Check <br><span class="smaller-title">A Ready-to-Use <br>Vital Signs Monitoring <br> Solution for Organizations</span> Vitalism Check <br><span class="smaller-title">A Ready-to-Use <br>Vital Signs Monitoring <br> Solution for Organizations</span>

rPPG pipeline

The following pipeline demonstrates an advanced video data processing that includes multiple techniques to extract a robust BVP signal.

Deep Learning Methods

We carried out measurements with deep learning methods using Pulse Rate Detection and UBFC-RPPG Datasets to compare our results, which is our main approach. We hope that deep learning will reduce error rates as a result of these measurements. DeepPyhs provides visualization of physiological information in videos using convolutional attention networks.The main purpose of PhysNet approach is using a Spatio-temporal network for rPPG signals from videos. Then it compares rPPG signals with Ground Truth ECG values

The Vitalism Check App

Eulerian Video Magnification (EVM)

Vitalism uses advanced computer vision and signal processing to monitor vital signs remotely. A key technique integrated into our system is Eulerian Video Magnification (EVM). EVM was developed by MIT in 2012. It magnifies small changes in videos.

For example, it can show your heartbeat by highlighting tiny color changes on your skin. EVM amplifies subtle color and motion changes in videos to extract physiological signals like heart rate and respiration. This method allows non-contact measurement of vital signs using a standard smartphone camera.

The process involves spatial decomposition, temporal filtering, and amplification of video frames to highlight imperceptible changes in skin tone due to blood flow. It works efficiently in real-time, making it practical for healthcare and telemedicine applications.

wave-border

Data Sets Used in Vitalism

UBFC: There are 42 videos in the UBFC dataset. Each person is about 1 meter away from the camera. Recorded at 30fps with a resolution of 640×480.

Pulse Rate Detection Dataset (PURE): consists of 10 persons (8 male, 2 female) recorded in 6 different setups. Reference: TU Ilmenau PURE Dataset .

PURE Dataset Request Step 1 PURE Dataset Request Step 2

We care about your users’ privacy

The application does not save images or input video streams used for measurement. Final measurement results are saved for historical purposes and user convenience. All results are secured, encrypted and accessible only by users and account admins with permission, which may be managed in account preferences.
Vitalism is fully committed to the privacy and security of our customers and users and operates in compliance with GDPR requirements.

Ready to try our Health Data Platform?

Ready to talk about your own specific use case for Vitalism technology? Let’s set up a call.

footer-wave-border

Download Vitalism App

Subscribe to our social media Pages

© Vitalism All rights reserved 2023