Time is a precious resource following a cancer diagnosis. Every year, millions of newly diagnosed patients wait weeks for next-generation sequencing results to determine whether they’re eligible for targeted therapies. During this critical window, the disease progresses and mutation-specific treatments remain inaccessible due to exorbitant costs and technical barriers. Entering this fraught landscape is Imagenomix, a startup founded by Dr. Matija Snuderl, Dr. Aristotelis Tsirigos and Dr. Daniel Orringer. All three are professors at NYU Grossman School of Medicine’s Department of Pathology. Using a combination of AI-driven molecular analysis and digital pathology, Imagenomix aims to bring greater speed, accessibility and precision to cancer diagnostics. 

How IGX Predict works

Imagenomix’s IGX Predict is a SaaS platform that uses AI to predict cancer mutations from pathology slides. It returns same-day actionable results, requires no new tissue samples and costs far less than traditional sequencing. “We discovered that you can make predictions from the way cancer cells are moving, aligning, clustering,” Snuderl explains. “Being able to do this is important because molecular testing is expensive and isn’t even available everywhere. That means a lot of patients cannot benefit from targeted therapy and precision medicine. So Imagenomix is about the idea that we can pull genomics from images and bring the promise of precision medicine to all cancer patients, regardless of location or economic status.” The platform installs like an app on digital pathology scanners and runs its proprietary algorithm on slides, surfacing the driver mutations used to guide targeted therapies. Currently, the platform features four product modules, each specializing in predicting driver mutations in lung, skin, brain and breast cancers. The use of AI cuts down on cost and turnaround time so significantly that the process wields the potential to majorly disrupt the cancer diagnostics market. 

From intuition to algorithm

Taking the research to market was on the team’s minds early on. The genesis of the idea was in 2017 when Snuderl was hit by a revelation - that his diagnostics work was a matter of pattern recognition.

“I ran molecular testing at NYU Langone while also being a pathologist who looked at slides so I was in the unique position of seeing what cancers look like and, weeks later, seeing their genomics,” Snuderl recalls. “I remember having a conversation with Aris Tsirigos predicting cancer mutations from slides I’d seen and he asked me how I knew. I said it’s because I’d seen so many of them. It was he who made the leap of faith that if Matija could make accurate predictions with his eyes then a deep learning model could be taught to find the patterns too.” 

After publishing the foundational work in Nature Medicine in 2018, the pair collaborated on forming a model that accurately predicted mutations in lung cancer. They soon brought in Orringer who helped them expand their scope across additional cancers. However, while their entrepreneurial spirit was strong out of the gate, it was initially slowed by a world that wasn’t quite ready for their innovation. “This was 2019, way before AI was hot,” Snuderl explains. “The papers made a huge splash and NYU introduced us to a few VCs but we didn’t get much interest then because everyone thought it was crazy - a fluke or artifacts or something.” The key was to keep that entrepreneurial engine revving and to wait for a better moment. “A few years on, Orringer said to me that it was now or never, we’ve got to do it,” Snuderl says. “It will change how cancer patients get treated. At that time, we honed down further on our models, a bunch of papers confirmed what we saw and then AI finally became hot and that’s what helped us make the leap from academic research to startup.” Imagenomix was incorporated in 2022 and it was off to the races. 

“Waiting for a better moment doesn’t mean waiting for a perfect time,” Snuderl clarifies. “For example, when we started Imagenomix, interest rates were skyrocketing and VC money was really drying out. On the other hand, we lucked out timing-wise in terms of AI and the technology for digitization advancing. There’s never perfect timing. It just happens and you do it.” Confidence in your concept also helps. “At one point, right before we launched the company, a bunch of digital pathology startups were coming up and I was wondering should we even be doing this,” Snuderl says. “I was saying to Dan that there’s just so many of them. And he said, ‘They’re all doing it wrong. That’s why we have to do it right!’”

Leveraging the NYU ecosystem

Doing it right involved learning some new skillsets. “I didn’t have any startup experience or know how these things went,” Snuderl says. “When I first came to NYU in 2013, somebody sent me downtown to meet Frank Rimalovski and visit the Leslie eLab and I loved it. I loved the environment, meeting the founders, and availing of the activities the team organized. They’ve built an incredible ecosystem for budding entrepreneurs who have an idea but don’t necessarily know how to start the next step.” Naturally, when the time came to get Imagenomix off the ground, Snuderl and his co-founders reached out to the NYU Entrepreneurial Institute team again. “It was the introductions, connections and support through the entire journey of launching and building Imagenomix that I found incredible. They were also invaluable in connecting us to other resources around NYU.” 

One such resource was NYU’s Technology Opportunities & Ventures office which the team worked with closely to file patents and license IP. “If you’re an affiliate, NYU has a really great system for protecting intellectual property,” Snuderl says. “They’re really proactive and take care of it all. You don’t have to be a lawyer or validate the cost. So when you build your startup, you know you’re protected. It’s a huge advantage.”

Bridging the molecular testing gap

Four years on from incorporation, Imagenomix is on a roll. In May 2026, they closed a $3.2M seed round led by Modi Ventures, with participation from NYU Langone Health, the NYU Innovation Venture Fund and Metavallon VC. They’ve integrated directly with Proscia, one of the world’s largest digital pathology platforms. They’ve grown steadily, building an able team led by CEO Travis Wold. And through it all, they remain focused on expanding access and partnering with pharmaceutical companies to improve clinical trial efficiency. “I see it as our mission that every single cancer patient should have access to molecular testing,” Snuderl says. “You can’t get targeted therapy without identifying the tumor’s mutations so it’s a paradox that insurance will pay for the drugs but not the testing that gets you to the drugs. I’ve coined the term ‘molecularly underserved communities’ to mean people who have cancer but can’t get access to testing. I’m hoping Imagenomix will bridge that gap completely whether you’re in Manhattan or South Dakota or really anywhere in the world.”