Photomix project to use photonics and AI to address diagnostic challenges

A newly-launched European research project uses photonics technology and artificial intelligence to address three global diagnostic challenges: non-invasive head & neck cancer screening; microplastic pollution detection in drinking water; and evaluating polymer biodegradation in the oceans. While these applications may seem unrelated on the surface, they share a fundamental sensing challenge: sometimes, examining a complex sample in only one way does not yield enough useful information.

The new technology aims to make testing faster, cheaper, and more accessible by using simple screening tools first, then bringing in more advanced instruments only when they are needed. These are among the challenges being addressed by PHOTOMIX. The team will explore whether the same underlying technology can also rapidly identify and characterize microplastics in drinking water and help chemical manufacturers assess how new polymers biodegrade in aquatic environments.

“Multimodal sensing itself is not a new concept,” said Dr Alessandro Giusti, R&D Director at the Cyprus Research and Innovation Center (CYRIC) and coordinator of PHOTOMIX. “Each sensing technique has its own advantages and disadvantages, and it can offer a different perspective on whatever you're trying to measure.”

“Researchers frequently combine different optical technologies so that the same sample can be examined from multiple angles. The real bottleneck has always been how to physically integrate them.”

One traditional approach is to integrate several sensing technologies onto the same chip or within the same instrument. However, fundamentally different optical systems do not make comfortable neighbors; trying to physically combine them can introduce immense mechanical, optical, and alignment challenges. Rather than attempting complex physical integration, the project keeps the individual instruments separate and combines their findings at the data level.

Five distinct sensing modalities – including FTIR spectroscopy, Raman imaging, visible imaging, and photothermal spectroscopy – will feed data into a common PHOTOMIX cloud platform. Advanced statistics, chemometrics, and artificial intelligence will then fuse and interpret the outputs.

“The main innovation is in the way we're trying to combine all the data on the same PHOTOMIX platform,” says Giusti.

Screening for cancer without biopsy

The planned approach would use an initial combination of photonic sensing modalities at the point of care to evaluate a lesion. If nothing concerning is detected, the approach could potentially help avoid an unnecessary invasive procedure. Only if the screening identifies something suspicious would the patient proceed to a biopsy, at which point a third, more sophisticated PHOTOMIX sensing modality would provide deep analysis.

PHOTOMIX’s second application examines exactly what is floating in our water. Detecting microplastics is currently a painstaking, laboratory-bound process. Water samples must be filtered and scanned micrometer by micrometer, a process that can take hours or even days. The aim here is to reverse that trend by developing portable, field-ready instrumentation, with the project's low-cost screening instrument targeted at around €10,000 ($11,000). Rather than scanning everything in maximum detail, a rapid first scan will identify specific regions of interest within minutes.

“You can already exclude areas of your filter that you don’t want to analyze in more depth,” said Giusti. “When you find a region of interest, you go there and run a more in-depth analysis.” Furthermore, the system will go beyond merely spotting small particles. By combining imaging with techniques that provide chemical signatures, PHOTOMIX aims to provide chemical characterization, determining not just that a particle is present, but also whether it is plastic and, potentially, what type of plastic it is.

Watching plastics disappear

The third demonstration addresses another side of the plastics challenge. As manufacturers race to develop sustainable polymers with a lower environmental impact, industry needs a reliable way to verify how these materials degrade. Here, PHOTOMIX will investigate whether its combined data platform can help the chemicals industry test polymer biodegradation rates directly in aquatic environments. Importantly, the project’s flexible data-fusion design means it does not require deploying all five sensing modalities for every problem.

By the end of the project, Giusti expects the complete platform to reach Technology Readiness Level (TRL) 6–7, indicating validation in a real-world environment. While full commercialization of the five-modality platform will require more time and funding, individual sensing modalities or dual-technology pairings could become commercially available sooner, demonstrating the platform’s value in practical, near-term stages.

For Giusti, that marks a significant shift from earlier research programs. Rather than asking whether the underlying photonic technologies can work, PHOTOMIX starts with technologies whose principals have already been demonstrated and asks how to bring them closer to real-world use.

“We know what we have. We know that the technologies work. We don't need to prove the concept again. We want to work on something higher that can eventually be closer to the market or to the application,” said Giusti.

PHOTOMIX will run until November 2029. Coordinated by the Cyprus Research and Innovation Center (CYRIC), the €5.96 million ($6.54 M) project brings together 12 research, academic and industry partners from Cyprus, France, Italy, Germany, Austria and Greece.