How Advanced Raman Imaging Could Transform Diagnostics
A broadband SRS image of human head and neck tumour tissue acquired with 10-ms pixel dwell time. Image copyright: Cambridge Raman Imaging Ltd.
By Dr Karen Steward
For decades, Professor Giulio Cerullo has worked at the forefront of optics and photonics, developing technologies that use ultrashort pulses of light to reveal the hidden properties of materials. As a physicist at Politecnico di Milano, his research has ranged from fundamental studies of light–matter interactions to practical applications with the potential to improve healthcare and industry.
One of the most exciting outcomes of this work is Cambridge Raman Imaging, a spin-off company co-founded by Cerullo and collaborators including Professor Andrea Ferrari, director of Cambridge Graphene Centre. The company has developed a unique imaging instrument capable of identifying the chemical composition of materials rapidly and without the need for traditional chemical processing. While its long-term ambition is to transform medical diagnostics, the technology is already attracting attention in areas ranging from pharmaceutical manufacturing to microplastics detection.
Karen Steward (KS): Can you tell us about your background and how this project began?
Giulio Cerullo (GC): “I'm a full professor of physics at Politecnico di Milano and I'm an experimentalist working on optics and photonics for more than 30 years.”
Cerullo explains that his collaboration with Professor Andrea Ferrari began through research on graphene and other two-dimensional materials. Around seven years ago, discussions about practical applications of graphene led to a new idea.
“We came up with an idea to simplify a new technique for imaging, which is called Raman imaging, where graphene could contribute to make this technique simpler and more effective.”
That idea eventually became Cambridge Raman Imaging. The company was founded in Cambridge in 2018 and later expanded to Milan, where many of Cerullo’s former students joined the team.
“We managed to fulfil our dreams. We managed to build it, and now it is working. It's already a commercial product, so several customers have already bought it.”
KS: What makes this imaging instrument unique?
GC: “What it can do is essentially tell you the chemical composition of a substance.”
Unlike conventional microscopes, which often rely on dyes or stains to reveal structures, this system identifies materials based on their molecular composition. This capability opens the door to applications where understanding chemistry is more important than simply seeing shapes or colours.
“Our first ambition, which is still our long-term dream, is to use this for medical diagnostics.”
Cerullo points out that the basic process used to analyse biopsies has changed surprisingly little over the past century. Tissue samples are typically stained with dyes and examined visually by highly trained specialists.
“It's quite subjective,” he says. “Essentially the histopathologist is trained after many years.”
His team's instrument takes a different approach.
“You take the tissue as it is and you put it in the machine and you get an image. And this image has the chemical composition in every point of the image.”
The technology can even recreate the familiar appearance of stained tissue without using any dyes.
“We can reproduce what we call virtual staining or AI staining. We can recreate the same colours that the doctors are used to, but without putting any dyes.”
A compact, turnkey all-fiber laser system developed by Cambridge Raman Imaging for SRS microscopy. Image copyright: Cambridge Raman Imaging Ltd.
KS: Why could this be particularly useful during surgery?
GC: “We could put our machine in the operating room and the medical doctor can take the piece of tissue, put it into the machine, and in maybe 10 minutes they get the image.”
This could be especially important during delicate procedures such as brain surgery, where surgeons need to know precisely where a tumour ends while preserving as much healthy tissue as possible.
Rather than waiting for laboratory analysis, doctors could receive near real-time information during the operation itself.
KS: How does artificial intelligence fit into the picture?
GC: “Now we have what we call a molecular avatar. We have digital data.”
Because the instrument produces digital chemical maps, the data is ideally suited for AI analysis.
“If you train an AI, then the AI can provide information on your sample,” Cerullo explains, while emphasising that human experts would still remain responsible for diagnosis.
He believes AI could help speed up assessments and potentially improve accuracy by reducing false negatives and false positives.
KS: What impact could this have on patients?
GC: “Sometimes people wait for weeks and it's not a nice wait.”
Beyond improving accuracy, Cerullo hopes the technology could significantly reduce the time required to analyse tissue samples.
“We hope that we can be more accurate, that we can be more precise in the diagnosis.”
Faster answers could reduce patient anxiety and allow treatment decisions to be made sooner.
KS: Are there applications beyond healthcare?
GC: “We realised that now we have potential that goes beyond the simple tissues.”
One major area of interest is pharmaceutical manufacturing. Drug manufacturers must verify that tablets contain the correct amount and distribution of active ingredients.
“With our machine we can take the tablets and test them within a few minutes. We can make an image tell what the percentage of pharmaceutical ingredient is, and how it is distributed.”
This could help manufacturers detect problems earlier, reducing waste and improving quality control.
“Another big application that we are looking into is microplastics.”
Microplastics have become a major environmental concern and identifying them accurately can be difficult.
“The plastics have characteristic molecular vibrations,” Cerullo explains. “We can really fingerprint them very accurately with our machine.”
By analysing filtered water samples, the system can rapidly identify and quantify different plastic particles.
“We talked to the environmental agencies and they were really impressed. They said this is really what we would need.”
Cerullo believes the technology could eventually become a standard tool for certifying water quality and monitoring environmental contamination.
KS: Could it help address antibiotic resistance?
GC: “One of the most dramatic health emergencies in the next maybe 20, 30 years will be antimicrobial resistance and our tool could definitely help here.”
The team is exploring whether the instrument can identify bacterial biofilms and determine which types of bacteria are present.
“If you have an infection, knowing what kind of bacteria they are is essential, so that you can try to get the right medicine.”
Rapid identification could help doctors select treatments more effectively and avoid unnecessary use of antibiotics.
KS: Why is the technology so versatile?
GC: “Wherever you want to identify something, our tool can be useful. Everything essentially is made of atoms, and these atoms vibrate, and we measure these vibrations in a simple way.”
The principle behind the instrument is surprisingly universal. Because every material has a unique molecular signature, the technique can potentially be applied wherever rapid identification is needed.
KS: Will these instruments eventually become portable?
GC: “At the moment, it's still a lab-based machine, but we have designed it to be usable by a nurse or technician, you don’t have to be an expert.”
While not yet handheld, the system has been designed with practical deployment in mind. Cerullo's team deliberately focused on robustness and ease of use. The design uses optical fibres rather than the complex arrangements of mirrors found in many advanced laser systems.
“We have designed it with portability idea in mind.”
KS: What is your ultimate goal for the technology?
GC: “Having something that does medical diagnostics would for me, also from a personal point of view, really show that my resources are doing something worthwhile.”
Although industrial applications may provide the fastest route to commercial success, Cerullo remains motivated by the possibility of improving healthcare.
If successful, the technology could help doctors make faster, more accurate decisions, while also finding valuable roles in environmental monitoring, pharmaceutical quality control and many other fields yet to be discovered.
About the interviewee
Giulio Cerullo is a full professor with the Physics Department, Politecnico di Milano, where he leads the Ultrafast Optical Spectroscopy laboratory. Prof. Cerullo’s research activity concerns, on the one hand, pushing our capabilities to generate and manipulate ultrashort light pulses, and on the other hand using such pulses to capture the dynamics of ultrafast events in molecular and solid-state systems. He has published over 560 papers, which have received >37000 citations (H-index: 96 on Scopus). He is a Fellow of the Optical Society of America, of the European Physical Society and of the Accademia dei Lincei and past chair of the Quantum Electronics and Optics Division of the European Physical Society. He has received two ERC grants (Advanced Grant in 2012 and Synergy Grant in 2025). He has been general chair of the conferences CLEO/Europe 2017, Ultrafast Phenomena 2018 and the International Conference on Raman Spectroscopy 2024. In 2023, he received the Quantum Electronics Prize of the European Physical Society. He is the co-founder of two spin off companies (NIREOS and Cambridge Raman Imaging).