Imperial students take heart health wearable to global IoT final

Five Imperial students reached the top 10 in the global final of the Silicon Labs IoT Challenge 2026 with a wearable prototype that aims to help earlier detection of heart failure.

Co-organised by FPT Software and Silicon Labs, the international engineering competition invited students from 140 universities to use the Internet of Things (IoT) and Edge AI to tackle real-world healthcare problems.

This year’s competition attracted 687 teams, with just ten teams selected to compete in the global final in Vietnam, with the Imperial team as the highest-placed UK entry.

Pranav Sharma

"Could we build something that might one day genuinely matter to someone living with heart failure?"

Just months after first putting an idea on a whiteboard at Imperial, we travelled to Hanoi with a working prototype of our device: ArterialAge.

Our interdisciplinary team brings together Abhi Kapoor and me – Electronic and Information Engineering students from EEE, Krish Thakrar from Mechanical Engineering, Anjali Ketan from Medical Biosciences with Management, and Joshika Manikandan from Chemistry.

Building for something bigger

ArterialAge began with a question: could we build something that might one day genuinely matter to someone living with heart failure?

By the time some patients hear the words ‘heart failure’, the disease has already changed their life and the lives of the people around them. We wanted to build something that could help change that.

Last May, we started meeting with Professor Kim Parker, Emeritus Professor of Physiological Fluid Mechanics in the Department of Bioengineering, and Dr Sonali Parbhoo, from the Department of Electrical and Electronic Engineering, to explore earlier identification of heart failure with preserved ejection fraction (HFpEF) outside of a hospital setting, before the condition becomes life-changing. HFpEF is a form of heart failure where the heart pumps normally but is too stiff to fill properly.

After countless whiteboard sessions, research papers and draft ideas, we landed on a concept inspired by Professor Parker: measuring arterial stiffness, a key biomarker for HFpEF. We wanted to explore whether a wearable device could use this information to help identify the condition earlier.

From whiteboard to wearable

The device is worn on the ear and measures pulse wave velocity of the pressure pulse wave through arteries downstream of the aorta. Pulse wave velocity is a key measureof arterial stiffness which is the upstream driver of HFpEF.

Turning a simple concept into a working prototype was much harder than we expected. We built every layer of ArterialAge ourselves: custom printed circuit boards (PCBs), CAD designs, optical shielding, signal processing pipelines and a machine learning model.

You build something, break it, rethink it and build again. Eventually, something that only existed in our heads became something real in front of us.

The result was our first physical ArterialAge prototype. Version one demonstrated a real, verified measurement.

We then successfully pitched at the semi-finals, earning one of the ten places in the global final.

ArterialAge on the global stage

Pitching wasn’t the hard part. Defending every design decision to experts was, and it’s where we learned the most.’ Abhi Kapoor

At the final, something we had built around workbenches at Imperial was suddenly sharing a stage with student healthcare innovations from around the world.

Taking something from an idea on a whiteboard to a physical device in front of experts on the other side of the world was surreal.

The judges brought decades of engineering and industry experience. Presenting ArterialAge meant not only demonstrating our prototype but also explaining the thinking behind every aspect of its design.

As Abhi reflected in his account of the competition: ‘Pitching wasn’t the hard part. Defending every design decision to experts was, and it’s where we learned the most.’

We finished as the highest-placed UK team and among the global top ten. The experience also gave us valuable feedback to take into the next stage of development.

The people behind the project

Dr Parbhoo and Professor Parker were instrumental in getting ArterialAge this far. Having two academics we respect and admire so deeply believe in five students meant more than I can put into words. They gave us their time, challenged our thinking and pushed us to make both the science and our ambition stronger.

Dr Parbhoo said: ‘What impressed me most about the students was their deep desire to drive change and clarity of vision of how this could help doctors improve detection of HFpEF. When they came to me to ask if I’d be willing to help build something with them, it reminded me very much of my own motivation for pursuing AI in medicine. There was no hesitation in my mind. The fact that they have this clarity so early in their undergraduate degrees is remarkable, as it’s not something many students at that stage have. For me, that is what makes this collaboration so meaningful: a shared commitment to helping clinicians recognise disease earlier and giving patients and their families a better chance.’

We are also grateful to Roland Zana and Shoaib Javed for their advice and mentorship throughout the project.

Imperial’s Department of Electrical and Electronic Engineering played a critical role in making the physical build possible, providing lab access, equipment and support when other facilities were closed. We especially want to thank Emma Rainbow, Vic Boddy, Alfie Williams and Jingbo Chen from EEE for their patience and practical advice.

Our thanks also go to Silicon Labs, FPT Software and the wider IoT Challenge team, particularly Rosie Duyen Nguyen Tran My and Chloe Quynh Anh Vu, for the hardware, mentorship and opportunity to present our work on a global stage.

Back to the building

Hanoi was a milestone, not a finish line.

We are now working on the second version of ArterialAge, incorporating feedback from the judges with a view to taking it forward as a real product. Our plans include a smaller, more practical wearable, refined signal processing, collecting more data and strengthening our machine learning model.

We also want to continue conversations with GPs and other healthcare professionals to understand how the technology might be developed and evaluated for use in clinical practice.

We came home incredibly proud of what we had built, but also knowing how much more there is to do. For now, the five of us are back at Imperial, building and testing, with the same goal in mind: to create something genuinely useful and to turn what we learned in Hanoi into the next version of ArterialAge.