Engineering a Smarter Microscope to Detect Cancer Faster

There's a moment in almost every cancer surgery that decides how the rest of the patient's treatment will go: the moment the surgeon has to decide where to stop cutting.

Get it right, and the tumor is gone. Get it wrong, and the patient may need a second surgery, weeks later, to remove tissue that was missed the first time. Today, that decision leans on a process that hasn't changed much in decades. We think engineering can change it.

‍

The problem: knowing, in real time, where the tumor ends

‍

During oncological surgery, one of the most delicate steps is defining the margin: how far to cut to be sure all diseased tissue is gone, without removing more healthy tissue than necessary.

Right now, that call depends on intraoperative frozen-section biopsies. A sample is removed during surgery, prepared and stained, and examined under a microscope by an anatomical pathologist. The pathologist visually assesses the tissue based on their expertise and the features revealed by the staining. The process is manual and can take valuable time during surgery, while the information available is limited to the sampled tissue. 

The consequence shows up in the numbers. Between 10% and 30% of some oncological surgeries end up requiring a second intervention due to positive margins, meaning that not all the diseased tissue was removed. That means a second surgery, more time under anesthesia, more cost for hospitals and insurers, and in procedures like kidney surgery, more time with blood flow interrupted, which directly damages the healthy tissue left behind.

It's a problem of information. Surgeons and anatomical pathologists are making a critical call without being able to see, at the cellular level and in real time, what's actually happening in the tissue in front of them.

‍

DIVER Technology

‍

Here is where DIVER comes in: a deep-tissue, multiphoton scanning microscope designed to reveal molecular information in tissue that conventional diagnostic methods cannot readily capture. 

DIVER uses a label-free contrast based on the molecular fingerprints of cells, rather than relying on chemical stains or markers. A tissue sample is still obtained during surgery, but instead of going through the conventional staining and histology workflow, it can be imaged directly with DIVER. The system captures molecular information from the tissue in real time, with the potential to support faster and more quantitative assessment of tumor margins. 

‍

‍

In practice, that means a technology like DIVER could eventually let a surgeon or pathologist look directly at tissue metabolism during a procedure, instead of waiting for a stained sample to come back from the lab.

DIVER’s technology was developed a decade ago at the University of California, Irvine, and it's protected by two UCI patents. What's new is where it's headed next. Today, only a few instruments of this kind exist in the world: four in the US, and one at the Institut Pasteur de Montevideo, in Uruguay. 

That's where this story turns into an engineering story.

‍

Building the first clinic-ready DIVER microscope

‍

The scientists who brought DIVER to Uruguay

‍

Leonel Malacrida spent his postdoctoral years (2015 to 2019) working directly with DIVER's original inventors (Enrico Gratton & Alexander Dvornikov) at UC Irvine, learning to build, operate, and refine the instrument. He brought that knowledge back to Uruguay and built a functional replica at the Institut Pasteur de Montevideo. Today, he directs the Advanced Bioimaging Unit (UBA). Luis Inchausti, precision machinist at Institut Pasteur de Montevideo with 40th yeras of experience, helps Leonel to build the first DIVER outside US. The UBA now operates around 15 microscopes and trains new scientists in advanced imaging.

That replica proved the science worked. What it couldn't do yet was leave the lab.

‍

Why Focus got involved

‍

A working DIVER microscope, as it exists in a research lab, is closer to a piece of scientific apparatus than a clinical instrument: an optical table the size of a room, weighing hundreds of kilos, operated by researchers who understand every one of its parameters. That's an extraordinary tool for a lab. But, it's not something you can wheel into an operating room.

In 2025, Focus partnered with Institut Pasteur to close that gap. That partnership became InDepth Focus, a startup backed by Lab+ Company Builder, which has since raised USD 750,000 in funding. 

Focus's role is specifically the engineering side of that equation: hardware, software, and firmware development, product design for manufacturing, and medical device expertise. We bring the discipline of turning a working prototype into a manufacturable, reliable medical instrument.

‍

As Javier Schandy, Focus's CEO, puts it:

‍

"Today the DIVER microscope can only be used by a handful of people, because it requires very advanced knowledge and has a huge number of parameters. One of our main challenges is usability: we're planning to lean on artificial intelligence so the system learns from past experiments, and the user can simply describe what they want to see."

‍

The impact: why this matters beyond the lab

‍

For DIVER to move from a research instrument to a technology that can be used in clinical settings, two challenges need to be solved: portability and usability.

The first is portability. DIVER's imaging capabilities depend on a sophisticated optical and detection system designed to capture signals from deep within tissue. Today, that technology is optimized for research environments. The challenge is to engineer a more compact, modular system without compromising the performance that makes DIVER valuable in the first place.

The engineering challenge goes beyond size. DIVER relies on measuring and generating signals on timescales ranging from nanoseconds to microseconds, requiring precise synchronization between its optical, detection, and electronic components. Integrating these systems into a smaller, portable platform while maintaining that level of timing precision is a key part of turning the technology into a clinical instrument.

The second is usability. DIVER generates rich imaging data, but turning that information into something clinically useful requires more than simply capturing better images. The system needs to help users interpret those images efficiently, even without extensive microscopy expertise.

That's where AI comes in. The roadmap includes integrating AI for automated image analysis, with the goal of improving efficiency in diagnostic processes such as tumor-margin assessment and making the system easier to use.

Put together, the goal isn't simply to make DIVER smaller or easier to use. It's to engineer a research-grade imaging technology into a practical clinical tool. One that can provide faster, quantitative information from tissue and ultimately support better decisions during cancer surgery.

‍

This is the kind of problem we like to work on

‍

Turning a groundbreaking piece of science into a device that survives daily use in a hospital is, at its core, a hard engineering problem. It's the kind of challenge Focus was built for.

If you're working on a product that needs to go from working prototype to manufacturable product, get in touch with our team. We'd like to hear about it.

‍

Contributors:

• Leonel Malacrida, Principal Investigator at Institut Pasteur
• Diego Bouvier, Engineering Manager at Focus