A unique insight into IntelliProve's technology
Brecht Dhuyvetters, CTO and co-founder of IntelliProve, gives a short overview of the algorithm developed by the company.
It is no secret that there is a huge workload problem in the medical sector nowadays. Healthcare providers are confronted with excessive amounts of work due the shortage in personnel. Next to an increased risk of burnout and psychological distress for these people, the enormous workload is also impacting the quality of care. In the long run, the IntelliProve technology has the potential to tackle these problems through video-based health monitoring. Time for a quick overview of our technology, written by our CTO.
A unique processing engine is at the heart of IntelliProve
The IntelliProve engine makes use of transdermal imaging technology. Yes, basically looking through your skin. The technique is based on photoplethysmography (PPG), which is an optical method to measure cardiac-synchronous blood volume change in body extremities, such as the face, fingers or earlobes. As the heart pumps blood, the volume of blood in the arteries and capillaries changes by a small amount, in sync with the cardiac cycle. The change in blood volume in the blood vessels underneath the skin leads to small change in the skin color (visible in the Red/Green/Blue spectrum), from which a PPG waveform can be estimated (see figure below). Vital parameters can be derived from a well-acquired PPG waveform using our innovative signal processing and machine learning techniques, trained by tens of thousands of PPG signals.
To extract the skin color changes that are actually related with the beating of the heart, we’ve developed cutting-edge algorithms that are able to determine facial regions with a high blood perfusion. We basically make a mesh of the face (see figure at the top) and analyze each region’s blood flow in real-time. A qualitative region will typically be one where the skin is not too thick, highly blood perfused and not covered with hair or other disturbances, such that the blood flow is nicely visible. This personalised computer vision approach enables us to aquire high quality PPG waveforms in different clinical conditions.
From bench to clinic
Using this high-tech optical technology, extensively tested through comparative studies using reference medical devices, we can give healthcare providers an easy-to-use software tool to measure vital signs of patients, with one click on a smartphone. Health parameters, such as heart rate or blood pressure, are measured using a mobile/smart device camera, simply pointed at the person’s face. I believe the main strengths of our product are the speed, contactless nature, and the fact that it can be used on any smartphone, tablet or laptop making it a very user-friendly – and easy to deploy. As we’re working on a connected smartphone platform anyways, results can automatically be transferred to the electronic health record – no risk of errors due to manually inputting data (yes, it happens!).
So, how does one start using our product, you may ask. Well, we’ve built our product such that it can be easily integrated into existing platforms, without the need to install yet another app or software – let alone another piece of hardware.
- Our product is a SaaS (Software as a Service) with an API endpoint.
- Integration with an existing platform is simply a matter of setting up the communication between the platform and our API.
- The optical measurement is recorded on device and that’s really the only requirement: the user’s device needs to be a smartphone or tablet with a camera. It’s the only sensor we need.
- After that, it’s simply a matter of calling our API and sending over the measurement data – which happens over an encrypted connection, of course.
- We process the data, extract the subject’s vital parameters, and send back the outcome to the device. No patient data needs to be saved on our end.
The digital revolution in healthcare
To wrap up this post, a final word on the direction the healthcare sector is heading to and how IntelliProve is jumping on that bandwagon.
With the Corona pandemic, it’s clear that the need for a much more preventive, personalised and decentralized healthcare system is there, and patients/healthcare providers will benefit from it. Take teleconsultations as an example. They are a safe and effective way to assess suspected health issues or guide the patient’s diagnosis and treatment, minimizing the risk of disease transmission. Teleconsultations also provide a much easier way for patients to do regular check-up of their health, or organize a remote follow-up after a hospitalization, e.g. to assess if the recovery of the patient is going well – in this way contributing to a more patient-centered healthcare system. IntelliProve perfectly fits in this digital revolution: integration of our technology into a teleconsultation platform would allow medical professionals to provide a more informed diagnosis, during the teleconsultation itself.
Furthermore, our remote health monitoring feature integrated into a health monitoring app, allows patients to assess their vital signs regularly and easily, allowing them to create an understanding about their personal health. When there are clear fluctuations, the user can take action immediately – leading to more successful treatments.
Our technology helps pave the way to digital biopsies, using our state-of-the-art computer vision technologies, to eventually contribute to a more efficient and better healthcare system.
That’s it for now. This post was a general overview of our technology and how it fits in the current and future healthcare landscape. Reach out to us if you want to know more about the technology, or hit me up on LinkedIn for a chat about the tools and techniques we use. Happy to do some knowledge sharing!