Leaving devices behind for scalable, remote health monitoring

Healthcare relies on physical hardware - blood pressure cuffs, wearables, and home kits. While these tools play a vital role in patient care, they also have built-in limitations. These include barriers of cost, access, and fragmented data, as well as limitations on who can be monitored, where monitoring is possible, and how consistently health can be tracked. 

Overcoming these challenges was vital because their impact is not purely technical. They determine human life, from patient outcomes to burnout in clinicians and the very constrained potential of remote care. 

If we could have a future where health monitoring is seamlessly integrated into everyday technology like smartphones and mirrors, it would make monitoring simpler for the user and more powerful for the provider than we can even imagine. Health insights would be as readily available as taking a photo. 

Challenges with traditional health monitoring

One challenge in remote health monitoring is data fragmentation. Information collected from different devices often gets stored in different places, making it difficult for healthcare providers and insurers to get a holistic and comprehensive view of a patient's health status. This lack of unified data leaks into ineffective program management and backs up patient misunderstanding over time.

Since traditional devices are uncomfortable, clumsy, or difficult to operate, patients simply don't use them often enough. This inconsistency leads to sporadic data collection. For providers offering remote monitoring services, this lack of high-quality and irregular information leads to misdiagnosis and cover hurdles for patients with their insurance.

To comprehend the scale of this issue, let's name some of these devices. In a traditional remote health monitoring setup, a patient receives a kit containing a blood pressure cuff, a wearable for continuous monitoring, a pulse oximeter, and/or a glucose meter and scale. Each requires shipping, patient education, connectivity to transmit data, individual device set up, ongoing maintenance and technical support, and monitoring for multiple devices per case.

Crucially, this imposes logistical constraints on reach, speed, and scalability. Expanding remote monitoring programs becomes too capital-intensive and operationally complicated. And it is too clumsy to scale from pilot programs to widespread deployment supporting thousands or millions of users.

As Dr. Digital CEO Ondrej Svoboda,said: 

"We realized we could take care of more people if we let go of the hardware."

‍

During the webinar, Przemek Jaworski, CTO of Shen AI, further emphasized the integration burden, stating that connecting data from "many various manufacturers of these devices is an immense task". One that adds enormous cost and operational strain on healthcare organizations, ultimately limiting the number of people that can be helped. 

But the revolution is here, and the tool for unlocking vital health insights only requires the device already in your hand. Device-free health monitoring is a fundamental leap forward in patient care and data collection, not just a theoretical idea.

‍

What makes Shen AI different?

Shen AI offers a fundamentally different approach to remote health monitoring. Instead of relying on physical gadgets, it uses advanced AI. It measures over 30 health markers, using a simple, 60-second facial scan.

Including:

• Blood Pressure
• Heart Rate
• Respiratory Rate (Breathing Rate)
• Heart Rate Variability (HRV / SDNN)
• Body Mass Index (BMI)
• Vascular Age
• Cardiovascular Event Risk
• Coronary Heart Disease Risk
• Stroke Risk
• Parasympathetic Activity
• Cardiac Workload
• Heart Failure Risk
• Coronary Death Risk
  

‍

The power of device-free health monitoring in action is that it works in real time using a standard smartphone or computer camera.

One of the biggest advantages is the ease of integration. As Jaworski mentioned in the webinar, “Even Bluetooth-enabled devices require costly integrations across vendors.” Shen AI simplifies this with one unified path for data access. It can be easily embedded directly into existing web or mobile applications with a simple SDK/API for this. This is a game-changer compared to the costly and complex task of integrating data from many device manufacturers. 

And we are only just getting started. During 2025, we are looking to add SpO2, HbA1c, High cholesterol risk, Anemia risk, and Hypertriglyceridemia risk metrics.  

For healthcare providers, that means a more comprehensive health assessment than with traditional hardware, better screening, risk stratification, and remote management of a wider range of conditions, all without needing additional devices. This expanded capability enhances what telehealth platforms offer their clients, keeping them competitive.

‍

Accuracy and simplicity at scale

Accuracy isn't a feature in medical care. It's literally life and death. The accuracy and simplicity is a hallmark of our device-free health monitoring offering at Shen AI.

To be accurate, we built our technology on rigorous scientific research and a massive, diverse dataset. 

As Przemek Jaworski detailed in the webinar:

"The first dataset we used was 70,000 people scanned with high-resolution videos, and then checked against biosignals collected by medical devices. The other is a dataset approaching half a million people right now, collected from the entire world. So the key here is diversity." 

‍

This extensive and diverse training ensures the technology works reliably across various skin tones, lighting conditions, environments, making vital, equitable healthcare access not only possible, but easy.

This commitment to precision is proven in our clinical studies. For instance, Heart Rate accuracy shows a Mean Absolute Error (MAE) as low as 0.1 bpm over 60 seconds and 0.2 bpm over 10 seconds, while Systolic Blood Pressure MAE is 8.57 mmHg and Diastolic Blood Pressure MAE is 5.78 mmHg. Our pursuit of precision continues with ongoing clinical validation.

Shen AI is the culmination of millions of heartbeats collectively. The technology's ability to process video in real-time directly on the user's device is key. This edge processing means patient data stays private and secure by design. No video or biosignal information is transferred or stored off the device during the scan. Everything happens locally. This focus on privacy and on-device processing makes AI health monitoring without devices a secure and appealing option for businesses and patients while addressing crucial compliance needs.

Shen AI offers an SDK, which gives developers one simple integration path. This contrasts sharply with the need to integrate with multiple vendors for traditional devices. This simplicity speeds up implementation, allowing providers to deploy remote monitoring faster. And since the technology runs on both web and mobile platforms, it's the lowest barrier to entry. 

‍

Next-gen technology: Multimodal Sensing

So, what sorcery powers this ultra-reliable, scalable remote health monitoring?

It's built using cutting-edge, robust Multimodal Sensing. This technology combines two different methods to analyze the video feed from a standard camera. It combines Remote photoplethysmography (rPPG) and remote ballistocardiography (rBCG) techniques. Think of them as a set of eyes.

rPPG analyzes subtle color changes in the skin, which are caused by blood flow as the heart pumps. It's a common technique, but it can be affected by external factors like ambient light and skin tones. As Przemek Jaworski explained in the webinar, rPPG works by using the camera's ability to detect these slight color variations: "Remote photoplethysmography is the way to utilize a video camera with red, green, and blue colors, which are different wavelengths, to extract the slight changes of color in skin, see the blood volume that changes, and how it flows through the skin because the skin is transparent".

Remote ballistocardiography is the second method that detects tiny micro-movements in the face. These micro-movements are caused by the heart ejecting blood with each beat. This method is less affected by lighting or skin tone variations. Using both rPPG and rBCG, Shen AI gets a complete picture of the heartbeat. Like seeing through both eyes instead of one, using both methods increases the measurement stability and accuracy. This is especially true in real-world situations where fluid conditions make device-free health monitoring more reliable.

So, how do you know which method to use when? Well, the system uses dynamic signal switching. It can choose the best quality input from either rPPG or rBCG at any given moment. This adaptability ensures the most reliable reading is captured. This adds crucial reliability for use cases where accuracy is paramount.

All advanced processing and analysis happen locally on the user's smartphone or computer. This preserves data security and ensures responsiveness. No sensitive video data leaves the device.

‍

Real-world impact: Scalability and improved outcomes

Moving past traditional devices unlocks significant benefits and applications. This scalable device-free health monitoring approach supports many use cases in modern healthcare.

• Telehealth & triage: Quick vital checks can inform clinical decisions. This can happen before or during virtual visits. Providers can access crucial data without needing patients to use separate devices.
• Chronic conditions management: Patients can easily conduct daily or weekly check-ins. They do not need to own or manage specific devices for remote patient monitoring without devices. This simplifies adherence to monitoring plans for conditions like hypertension or diabetes.
• Insurance & population health: This technology enables broad reach and access to health data across large groups. It allows insurers and public health programs to monitor populations without the logistical challenge of distributing hardware.
• Preventive health & home care: It makes regular health checks simple and accessible. It can become a seamless part of a daily routine for individuals managing their health at home.

As Remi noted, while traditional methods require sending device kits to each person, limiting scalability, a digital approach means:

"You can scale from one to a million in one day. And there's no limitation".

• Accessibility and convenience: Checking vitals is incredibly easy because it fits into anyone's routine using only the device they have.
• Reduced costs: Shen AI's digital approach is far more economical at scale than physical devices. It costs as little as $5 per patient per month, compared to traditional kits, which cost around $100 per month. This cost reduction makes remote monitoring at scale far more viable for everyone.
• Enhanced patient engagement: The ease of use leads to higher participation and more consistent data collection. Patients are more likely to monitor their health when it's simple and convenient.
• Scalability and broader reach: Access is no longer limited by device availability or logistics. The technology can reach patients who can’t access traditional care or devices.
• Improved health outcomes: Better data means better monitoring and more informed interventions by healthcare providers, such as early detection and proactive management.

‍

Conclusion

Shen AI's goal is clear: Empower every individual with accessible health information, use their everyday technology, and make complete health checks as simple as a daily routine.

The days of expensive, clumsy health monitoring hardware are over. Our tech is more than simple health checks; it's about unlocking truly scalable medical care. By leaving the limitations of the past behind, we start to see the possibility of reaching everyone, everywhere. Think about improving health outcomes on a global scale.  Think of fundamentally transforming the future of medical care. 

Watch the full webinar recording here: Leaving expensive devices behind for scalable, remote health monitoring.