TECH

A $10 sensor promises a medical laboratory on your wrist

A new light sensor, which cost only $10 to manufacture in a laboratory setting, could bring lab-grade precision to wearable health monitors. This would lower the cost of smartbands and smartwatches while overcoming one of their major shortcomings: a level of accuracy that has historically fallen short of medical-grade equipment.

The technology could also transform food testing, industrial quality control, and environmental monitoring—and even allow individuals to track health indicators, such as blood glucose levels, without the need for needles or periodic trips to a laboratory for testing.

This new type of optical spectrometer combines exceptional precision with a remarkably small footprint and low production costs. Measuring only a few centimeters in diameter, the sensor can be integrated into portable electronics and wearable health technologies.

Optical spectrometers work by analyzing light to reveal information about the substances with which it interacts. They are essential tools in medicine, industry, chemical testing, and environmental research. However, miniaturizing these instruments has traditionally meant compromising their performance. Compact spectrometers often feature limited measurement ranges, reduced resolution, or lower accuracy—making them better suited for identifying substances than for taking precise measurements.

The researchers believe their new design can overcome these limitations. Known as a "convolutional spectrometer," it employs a mathematical principle called the convolution theorem to analyze light in a fundamentally different way. According to the team, this approach delivers superior performance compared to existing types of spectrometers—including those based on light dispersion, Fourier transforms, or computational reconstruction—across several key areas. 

Traditional spectrometers typically analyze light by dispersing it or reconstructing spectra through computational methods. The convolutional spectrometer takes a different approach by performing the convolution process directly on incoming light.

It uses a sequence of optical components with periodic spectral responses, including unbalanced Mach–Zehnder interferometers and microring resonators. By adjusting these components proportionally, the system shifts its spectral response in a controlled way, allowing accurate spectrum recovery through fast Fourier transforms.

“The key insight was to go back to the mathematics and ask whether there was a fundamentally cleaner way to retrieve spectra,” said Dr. Chunhui Yao, a lead author of the study. “By using the convolution theorem directly in the optical domain, we avoid many of the limitations that have held miniaturized spectrometers back. This gives us high precision, strong noise tolerance, and very low computational overhead, all in a compact and low-cost system.”

Built on a silicon nitride photonic integration platform and combined with onboard electronics, the device operates across a broad near-infrared wavelength range (1200–1700 nm). It can capture and process data in less than a second. Its periodic design also enables almost unlimited expansion of spectral bandwidth without modifying the hardware. Resolution can be increased dramatically by adding more components to the system.

The researchers tested the spectrometer in a variety of practical applications. For materials and food analysis, it correctly identified plastics, pharmaceuticals, coffee, flour, and tea with a 100% success rate. It also measured concentrations in aqueous and organic solutions with accuracy around 0.01%, outperforming commercially available benchtop spectrometers.

Glucose and other parameters, without needle pricks...In tests described by the university, the sensor was also used to measure other biological indicators, such as blood lactate, blood alcohol levels, and skin moisture. These indicators, known as biomarkers, are characteristics or substances that provide information about how the body is functioning.

The ability to track glucose levels without piercing the skin is particularly significant for people with diabetes, who must monitor their blood sugar regularly. Currently, such monitoring often requires small skin punctures or the use of sensors inserted beneath the skin. However, the demonstration provided by the researchers does not prove that the device is already capable of replacing these methods. Long-term monitoring without blood sampling was carried out on a single participant, and the sensor’s performance will still need to be evaluated in larger groups and under different conditions of use.

“These biomedical demonstrations are particularly exciting,” notes Qixiang Cheng, the professor who led the study. According to him, one of the breakthroughs of this work was transforming a mathematical concept into a functional system capable of operating in real-world conditions.

The sensor outperformed large, expensive laboratory equipment in some tests. [Image: University of Cambridge]

How the sensor works...The device is an optical spectrometer—an instrument that analyzes the interaction between light and various substances to identify materials or estimate their composition. While this technique is used in laboratories, industrial settings, and environmental research, traditional equipment tends to be expensive and bulky.

Generally, reducing the size of such devices comes at the cost of precision or measurement capability. To overcome this, the team developed a model called a "convolutional spectrometer." Instead of separating light into its individual wavelengths, it uses a sequence of optical components and applies mathematical calculations to the resulting signals.

In this way, the device can retrieve information about the substance being analyzed without requiring a large system or intensive computational processing. "The key breakthrough was going back to the mathematics and asking whether there was a fundamentally simpler way to obtain spectra," says Chunhui Yao, one of the study’s lead authors, in a statement.

According to the researcher, this new approach helps overcome the limitations of miniaturized spectrometers by combining precision and resistance to interference with low cost. Built using silicon nitride—a material that allows light to be guided through tiny circuits—the device measures only a few centimeters in diameter and processes each measurement in less than a second.

One of the most notable demonstrations involved noninvasive monitoring of human biomarkers under realistic physiological conditions. The device accurately measured skin moisture, blood alcohol, blood lactate, and blood glucose levels. Researchers also successfully tracked glucose levels over extended periods in a single participant.

“These biomedical demonstrations are particularly exciting,” said Prof. Qixiang Cheng, who led the project. “What makes this work stand out is not just performance in the lab, but technical readiness. Dr. Chunhui Yao’s contribution was crucial in translating a mathematical concept into a fully packaged, robust system that operates reliably across temperature extremes and real-world conditions. That combination is what opens the door to practical deployment.”

The device maintained stable performance at temperatures ranging from –20°C to 80°C (–4°F to 176°F). This level of durability is uncommon among miniaturized spectrometers and is critical for wearable, industrial, and outdoor applications.

Beyond its measurement capabilities, the convolutional spectrometer offers a simpler design and lower computational demands than many competing systems. Existing high-performance miniaturized spectrometers often depend on complicated calibration procedures and computationally intensive reconstruction methods that can be vulnerable to noise. The linear nature of convolution allows this system to recover spectral information quickly and reliably while requiring very little processing power.

“This is a beautiful example of how elegant engineering can unlock real impact,” said Prof. Richard Penty, who contributed to the photonic system design and integration. “The architecture is simple, scalable, and manufacturable, yet it delivers precision that rivals or exceeds much larger instruments. Dr. Chunhui Yao played a central role in bringing together the photonic design, system integration, and experimental validation that made this possible.”

Surpassing laboratory equipment...To evaluate the device, the researchers tested it on a variety of materials and substances. The results demonstrated its potential for applications that typically rely on larger, more expensive laboratory instruments.

When used to identify plastics, pharmaceuticals, coffee, flour, and tea, the spectrometer achieved a 100% classification success rate. It also measured concentrations in aqueous and organic solutions with an accuracy of approximately 0.01%, outperforming commercial benchtop spectrometers.

In healthcare demonstrations—conducted under conditions designed to reflect real-world physiological environments—the device accurately measured various human biomarkers, including skin moisture, blood alcohol, blood lactate, and blood glucose.

"These biomedical demonstrations are particularly exciting. What makes this work exceptional is not just its laboratory performance, but also its level of technical readiness," said Professor Chunhui Yao of the University of Cambridge, UK. The team has already established a company to commercialize the technology.

Researchers from the University of Cambridge and the startup GlitterinTech have developed a $10 miniaturized optical spectrometer that can be integrated into wearable devices (such as smartwatches).

Key features:

• Technology: Based on a silicon nitride (photonic platform) that operates in the near-infrared wavelength range. It uses an optical convolution concept to analyze light and identify the chemical composition of substances.

• Cost and Size: It costs approximately $10 and measures only a few centimeters.

• Health Applications: It offers the potential for real-time, needle-free (non-invasive) monitoring of markers such as blood glucose, alcohol, lactate, and skin moisture.

• Other Uses: It can classify plastics, pharmaceuticals, coffee, flour, and tea with high precision.

Current status and limitations:

• Initial Tests: In laboratory settings, the chip demonstrated high accuracy in identifying materials. However, non-invasive glucose monitoring has been tested on only a single participant.

• Validation: Further clinical studies and large-scale testing are required before the product can reach the consumer market.

 

mundophone