ΑΙhub.org
 

New device monitors blood glucose levels using radar and AI


by
15 October 2020



share this:
Researchers test a prototype of a new diabetes device for prick-free glucose monitoring.

New technology can quickly and accurately monitor glucose levels in people with diabetes without painful finger pricks to draw blood. A palm-sized device developed by researchers at the University of Waterloo uses radar and artificial intelligence (AI) to non-invasively read blood inside the human body.

“The key advantage is simply no pricking,” said George Shaker, an engineering professor at Waterloo. “That is extremely important for a lot of people, especially elderly people with very sensitive skin and children who require multiple tests throughout the day.”

About the same size as existing glucometers, the rectangular device works by sending radio waves through the skin and into blood vessels when users place the tip of their finger on a touchpad. The waves are then reflected back to the device for signal processing and analysis by a machine learning algorithm (principal component analysis), telling users within seconds whether their blood sugar has gone up, down or remained the same.

Changes are measured in relation to a baseline reading that would be obtained every few weeks with a glucometer or a laboratory blood test to ensure accuracy.

“Our safe, reusable, pain-free device would eliminate the need for implanted sensors, patches or devices that use chemical reactions or fluid transfer through the skin,” said Ala Eldin Omer, an engineering PhD student who led the project.

Researchers are now exploring commercialization of the inexpensive technology – they estimate the device would retail for less than $500 – and developing a wearable device similar to a smartwatch that would be on at all times.

“This finding paves the way for continuous monitoring,” said Shaker. “Given the current pace of progress, I expect the technology to be available in a wearable form within the next couple of years.”

Safieddin (Ali) Safavi-Naeini, also an engineering professor at Waterloo, said the science at the heart of the diabetes device potentially has several additional applications. “Since many ingredients of blood have distinct electromagnetic properties, the same technology could be extended to other types of blood analysis and medical diagnosis,” he said.

Much of the work was conducted at the Wireless Sensors and Devices Lab and the Centre for Intelligent Antenna and Radio Systems (CIARS) at Waterloo. Collaborators included several researchers at Sorbonne University in Paris.

Read the paper in full

Low‑cost portable microwave sensor for non‑invasive monitoring of blood glucose level: novel design utilizing a four‑cell CSRR hexagonal confguration
Ala Eldin Omer, George Shaker, Safeddin Safavi‑Naeini, Hamid Kokabi, Georges Alquié, Frédérique Deshours & Raed M. Shubair.




University of Waterloo

            AUAI is supported by:



Subscribe to AIhub newsletter on substack



Related posts :

monthly digest

AIhub monthly digest: October 2026 – neurosymbolic planning, time-series forecasting, and robots for the home

  30 Oct 2026
Welcome to our monthly digest, where you can catch up with AI research, events and news from the month past.
AI pioneers

From teaching machines to teaching people: an interview with Michael Littman

  29 Oct 2026
AI Pioneer Michael Littman reflects on the evolution of reinforcement learning, the power of science communication, and how generative AI is changing the way we teach and learn.

From lab to life: Why translating AI advances to the real world is a major challenge

  28 Oct 2026
Systems that perform well in controlled lab settings often fall short when deployed in homes, in hospitals or on city streets.

Researchers get AI ‘drunk’ to expose new cyber security risks in chatbots

Fine-tuning large language models using drunken text makes them more likely to leak secrets and answer harmful questions.

AI is creating jobs as well as erasing them – but how rewarding are they?

  26 Oct 2026
One area of increased employment is in people “cleaning up” after LLMs, or quality controlling them.

How AI changes human decision-making

  23 Oct 2026
AI can identify patterns that people miss and bring new information into decision-making. However, this does not automatically lead to better decisions.

Competing in RoboCup@Home and winning the 2026 best paper award: an interview with Raphael Memmesheimer

  22 Oct 2026
Hear more about Raphael's research and his involvement in the RoboCup@Home league.

AI is distorting archaeology and flattening Indigenous Knowledge

While AI can help archaeologists, it can also spread pseudoscientific claims and false histories, stripping Indigenous Knowledge of context, consent and community control.


↑


AUAI is supported by:







Subscribe to AIhub newsletter on substack




 















©2026.05 - Association for the Understanding of Artificial Intelligence