Five Imperial College London students have reached the global top ten of the Silicon Labs IoT Challenge 2026 with an ear-worn prototype designed to investigate earlier detection of a form of heart failure.
Their device, ArterialAge, measures pulse wave velocity, an indicator of arterial stiffness. The team is exploring whether this information could help identify heart failure with preserved ejection fraction (HFpEF) outside a hospital setting. In this condition, the heart retains its pumping function but is too stiff to fill properly.
Co-organised by FPT Software and Silicon Labs, the competition challenged students to apply the Internet of Things and Edge AI to healthcare problems. It attracted 687 teams, with ten selected for the global final in Hanoi, Vietnam. Imperial finished as the highest-placed UK entry.
The project brought together Electronic and Information Engineering students Abhi Kapoor and Pranav Sharma, Mechanical Engineering student Krish Thakrar, Anjali Ketan from Medical Biosciences with Management, and Chemistry student Joshika Manikandan. Professor Kim Parker and Dr Sonali Parbhoo supported the team as it developed the concept.
Advertisement
Turning the idea into working hardware required the students to build each part of the system themselves, including custom printed circuit boards, CAD designs, optical shielding, signal-processing pipelines and a machine-learning model. According to the team, the first prototype demonstrated a verified measurement, providing the basis for its competition entry.
The final tested more than the device’s performance. Students had to explain their design choices to judges with engineering and industry experience, connecting the physical build with the problem they hoped to address.
‘Pitching wasn’t the hard part. Defending every design decision to experts was, and it’s where we learned the most,’ Kapoor said.
That experience is now informing a second version of ArterialAge. The team plans to make the wearable smaller and more practical, refine its signal processing, collect additional data and strengthen the machine-learning model. It also intends to continue discussions with GPs and other healthcare professionals about how the technology could be developed and evaluated for clinical use.
For the students, the project has provided experience of the complete development process: defining a need, combining expertise from different disciplines, building and testing hardware, and responding to external scrutiny. The next stage will be to establish how the prototype’s measurements could contribute to a useful healthcare product.