IntelliProve's software makes use of photoplethysmography (PPG) to detect changes in blood volume in the face, induced by the rhythmic pulsations of the heart.
We track the subtle fluctuations in skin colour resulting from facial blood flow by monitoring the reflection of light. Cutting-edge signal processing techniques and in-house developed algorithms transform these minuscule variations into precise and reliable physiological signals. This provides insights into the body's cardiovascular activity, including heart rate and heart rate variability.
Our process begins with a setup check, during which we assess the measurement conditions to ensure optimal settings for capturing a high-quality recording. This step is crucial in guaranteeing the accuracy and reliability of our subsequent analyses.
An in-house developed machine learning model detects the skin areas within the captured images.
Leveraging the data obtained from our skin segmentation model, our system navigates through the complexities of pulse waves, allowing for precise identification and characterisation of blood flow-rich regions.
Photoplethymography is a technique to detect variations in blood volume by shining light onto the skin and measuring the amount of reflected light. It's the technique that is used by finger pulse oximeters.
IntelliProve elevates this approach by leveraging your smartphone camera as a sensor to monitor fluctuations in ambient light reflected from your face. This technique is what we call remote photoplethysmography (rPPG).
We focus on the blood flow-rich regions determined in the previous step and apply advanced signal processing techniques to extract physiological signals, including the blood volume pulse and respiratory signal. Our robust signal quality assessment algorithms ensure the highest signal quality possible.
The physiological signals obtained during rPPG analysis contain valuable information relating to the final biomarkers and cardiovascular characteristics of your body.
Our custom-built extraction pipeline, consisting of signal processing algorithms and ML models, uses these signals to determine parameters such as heart rate, respiratory rate and heart rate variability.
More than 1 million datapoints are actively used for testing and validating our models.
Percentage of heart rate measurements with an absolute error below 5 beats per minute.
Detection rate of low/medium and high mental health risk.
Detection rate of stressful situations through increased sympathetic activity of the autonomic nervous system.
Percentage of heart rate variability (SDNN) measurements with an absolute error below 20 milliseconds.
Detection rate of a resonant breathing state.
IntelliProve increases user engagement by 20% within the first 3 months.
Why wait? Let’s start a conversation.