A $10 Chip Smaller Than a Coin Could Read Blood Sugar Without a Needle

A new light sensor costing about $10 could one day bring laboratory-grade measurements to wearable devices, including a way to follow blood sugar without piercing the skin, according to ScienceDaily, which summarized work from the University of Cambridge. The researchers say the same technology could also be used to test food, check industrial quality and monitor the environment.

The device was developed by a team at Cambridge together with GlitterinTech, a start-up founded by the same research group. It is only centimeters across, small enough to fit into portable electronics. The glucose result, however, comes from a single participant, and the team says more testing is needed before it can become a medical product.

What this means for wearables and health tracking

Many people with diabetes still rely on finger-prick tests or sensors worn under the skin. The Cambridge team points to a possible alternative: a compact optical sensor that reads the body's chemistry from light alone. In their experiments, the device measured skin moisture, blood alcohol, blood lactate and blood glucose under conditions designed to resemble real physiological environments.

The researchers also tracked glucose over an extended period in one participant without invasive procedures. That is an early demonstration, not a clinical trial, and nothing in the report suggests a product is close to reaching shoppers.

Price is part of the story. Small spectrometers have usually forced a trade-off between size and performance, with limited measurement range, lower resolution or reduced accuracy. A $10 part that avoids that trade-off could make precise optical sensing practical in consumer gadgets, food-safety checks and factory lines that now depend on larger, costlier instruments.

How the sensor works

An optical spectrometer analyzes light to learn about the substances it has passed through or bounced off. Conventional designs either split light into its separate wavelengths or rely on heavy computation to rebuild the spectrum.

The Cambridge device, called a convolutional spectrometer, takes a different route. It applies a mathematical operation known as convolution directly to the incoming light, using a series of optical components whose responses repeat. Fast Fourier transforms then recover the spectrum. Dr. Chunhui Yao, a lead author, said the approach delivers high precision and strong noise tolerance with very little computing overhead.

The team built the chip on a silicon nitride photonic platform and combined it with onboard electronics. It works across near-infrared wavelengths from 1200-1700 nm and can collect and process a measurement in less than a second. Because the design uses repeating patterns, the researchers say its spectral range could in theory be extended almost without limit, and adding components in sequence could raise its resolution exponentially.

What the tests showed

The researchers tested the sensor on a range of materials. It identified plastics, pharmaceuticals, coffee, flour and tea with a 100% classification success rate in the reported tests. It measured concentrations in water-based and organic solutions with accuracy of approximately 0.01%, which the team says beat commercial benchtop spectrometers in those experiments.

The sensor also kept working between -20°C and 80°C, an unusually wide range for a miniature spectrometer. Prof. Qixiang Cheng, who led the project, said the work stands out for its technical readiness, not only its laboratory performance, because the system operates reliably in real-world conditions.

What remains unproven

The results come from the research group's own experiments, and the summary does not describe independent replication. The health demonstrations are promising but limited, and the glucose tracking involved a single person. Turning a lab prototype into an approved medical device would require substantially more testing, and no timeline was given.

Around Prague, based on reporting by ScienceDaily