Itai Hayut, is the CEO and Founder of Scopio Labs, a startup company developing a novel digital microscopy platform that uses advanced computational photography techniques in order to perform imaging and analysis of microscopy samples.
What initially attracted you to the world of biomedical research and innovation?
I studied physics at the Hebrew University, but I was always fascinated with human biology and medicine, so it was natural that I would eventually bring the two together in some way. One of my early projects was in the area of medical devices, which turned into a successful grant from the Israeli government. I think that was the pivotal point when I knew I wanted to apply technologies like computer vision and deep learning to push the boundaries of medicine and innovate in a way that would do some good in the world.
I spent a fascinating summer in Silicon Valley at Singularity University exploring exponential technologies and applying them to the world’s most urgent problems. This served as the perfect springboard for me into the convergence of AI and the medical field.
Could you share with us the genesis story behind Scopio Labs?
This story also started back when I was at university with my co-founder Erez Na’aman. We always dreamed of doing something together – leveraging IT and AI to make an impact on the world. During our time working in hospitals, we would watch practitioners look through traditional microscopes and count cells manually. It was stunning to us how much of the medical world still relied on diagnostics performed using the manual microscope, which is analog, outdated, and labor-intensive. What’s more, it offers no easy way to share microscopic data, collaborate with specialists, or perform the sort of AI-based image analysis that is commonplace in other industries.
We thought that surely that with our background in physics, IT and medical devices, we could bring microscopy and hematology specifically into the digital age. All we had to do was buy a scanner to scan the slides, and then we could apply the AI tools on top of that, right?
But when we started looking for a scanner, it didn’t exist. And that’s really where the story of Scopio began. Our first task was to build a scanner, and with it we could apply high quality data and advancements in deep learning to revolutionize hematology and transform disease detection and diagnosis.
Scopio Labs offers full-field Scanning at 100X resolution, what were some of the key technologies that enabled full-field scanning versus traditional narrow view options?
Overcoming the longstanding traditional trade-off between field of view (FOV) and resolution is the biggest breakthrough. Scopio’s revolutionary digital imaging platform captures large scan areas at 100X magnification without sacrificing the full field of view, and this is all made possible using advanced technology called computational photography.
In our fully digital solution, we use computational photography to permit a low-resolution objective to rapidly acquire digital images under different illumination conditions. To that we apply sophisticated algorithms, and a sharp, 100X-resolution, full-field image of the sample is computed. It all happens a lot faster and more efficiently.
Many of today’s digital solutions for analyzing peripheral blood smears are only able to create a partial view resulting in single snapshots of cells – just single digital images. These digital solutions herald initial digitization and automation efforts to transform this manual industry, but are limited in their application for end-to-end in-depth digital analysis: with single-cell snapshots you do not get the full context of the slide. Or more importantly, areas of clinical interest such as the feathered edge are not included, which would require the lab technician to go back to the manual microscope to complete the case review
Digital technology up until now had limits to how many FOVs on the slide can be digitally analyzed. Current digital cell morphology imaging systems for PBS cannot achieve full 100X oil immersion resolution at scale and are limited to only providing snapshots of cells.
The limiting factor was time, because they use a high-resolution camera sensor mounted on a 100X magnification objective – the part of the scope closest to the object, every step takes forever. Things like precise objective movements, system stabilization, and focusing are all very time-consuming. And cost prohibitive for smaller and medium labs. Computational photography replaces expensive optics and mechanics, dramatically reducing the overall solution cost also while at the same time providing a completely digital full field cell morphology solution.
What are some of the benefits of full-field scanning?
When you have full-field visibility at high resolution, it essentially means you can both see the big picture in context and zoom in on any portion of it at 100X resolution. Imagine a satellite image of the entire Amazon rainforest with the ability to zoom in on any tree and leaf with absolute clarity.
For a lab practitioner or hematopathologist, this means you see everything there is to see with no gaps or blind spots. Whether the region of interest is located at the farthest edge of the blood smear, the feathered edge, you can pan to it and see it at 100X resolution. At the same time, you can look at the entire blood smear at once in full context. Both capabilities are vital for clinical decision-making and are capabilities that were not possible until now.
What’s more, semi-digital solutions automatically select an analysis area and retrieve only single snapshots – 1 tree out of hundreds if I go back to our rainforest example. So if a sub-optimal area is selected by the device, the tech has to go back to the manual microscope anyway, and they haven’t gained anything.
With our full-field cell morphology approach, the tech or hemapathologist can pan out and zoom in on any cell or group of cells anywhere in the digital image – so can a colleague at a computer screen hundreds of miles away.
Could you discuss some of the remote capabilities that are offered by this platform?
Remote access to full-field scanned images is a powerful advantage for our users. Offsite staff or collaborators can have the same access to the scans as anyone in the lab, with the same ability to see the entire slide and/or zoom in on any area of interest at 100X. A clinician, consulting hematopathologist or other party can review, consult and collaborate from any location.
The result is that diagnosis can happen much faster, second-opinion referrals can happen instantly, treatment can begin faster and with greater confidence, and physicians can improve patient outcomes.
Obviously remote access has been key during the pandemic because it allows for information sharing while minimizing staff interaction and specimen handling and processing.
One of the core business solutions that is offered is ScopioVet, what is this application specifically?
ScopioVet is a unique application of our technology that also incorporates remote pathology services. It enables veterinary clinics to scan and analyze cytological specimens at the point of care and get pathology results within an hour. So, for example, if a dog comes in with a lump, the clinician can take a fine needle aspiration, prep a slide, scan it with the ScopioVet X100 device in the clinic, then transmit it instantly for review by our global network of pathologists who are available 7/24/365.
Results come back within the hour, and the clinic can consult directly with the pathologist if desired. They can do this with all fine needle aspirates, peripheral blood smears, ear swabs and any other body fluids.
So instead of sending out the specimen to a path lab for review and waiting days for a result, the clinic has answers at the point of care before the client even leaves the office – so a treatment plan can be implemented right away. You can imagine the benefits this has in terms of patient outcomes, client satisfaction, clinic profitability, staff productivity – it has shown to even help with client compliance. It’s really a transformative advancement.
What’s your vision for the future of cell morphology platforms?
For cells to speak the truth, you must analyze thousands of cells at a time at a resolution high enough to spot the most minute details. Otherwise, results are inconsistent, diagnoses are inaccurate, and key findings are missed.
And yet, today, cell morphology at that scale is a complete fantasy. Analyzing hundreds of cells is already beyond human capacity, let alone thousands. Our very best tools are limiting by design – narrowing our field of view to individual cells at a time. And even today’s most advanced AI technologies restrict our analysis to snapshots of cells, giving us a small handful of puzzle pieces, when what we really need is the complete picture.
Now that we’ve cracked the code of full-field digital cell morphology, the possibilities for using AI to enhance human observation and diagnosis are virtually unlimited. Already our systems use machine learning to differentiate among the morphology of blood cells. In the future, AI will push us farther and farther beyond the limits of human capabilities by automating the analysis of tens of thousands of cells at a time – far more than any human can quantify. A breakthrough for accessing and analyzing all of the information morphology holds and enabling new diagnostic insights at the highest degree of accuracy, at a scale the world has never seen before. Detection and diagnosis of cancers, infections, diseases, and relapses earlier than was ever thought possible
Is there anything else that you would like to share about Scopio Labs?
I hope that when other people see our work and the technology that we’re putting forward, they get as excited about it as I do. I’m still fascinated by both the biology and the mechanics of the human body, and I never get tired of learning new things about cell morphology from our customers in the medical profession. I’m so proud that we are using AI in such a profound way: to empower humans to see things more clearly, find answers faster, and ultimately prevent suffering. We’ve only just opened the door, and there’s an entirely new frontier of innovation and possibility ahead for this company.
Thank you for the great interview, I look forward to following the progress of this groundbreaking technology, readers who wish to learn more should visit Scopio Labs.
- Anastasia Leng, Founder & CEO of Buzz Solutions – Interview Series
- ‘Smart’ Walking Stick Helps Visually Impaired Grocery Shop
- What is a Data Analyst? Salary, Responsibilities, Skills, & Career Path
- 10 Best AI Random Face Generators (January 2023)
- Scientists Combine Conventional Robotics and Microfluids