Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
If you’ve ever had a PET scan, you know it’s an ordeal. The scans help doctors detect cancer and track its spread, but the procedure itself is a logistical nightmare for patients.
It starts with fasting for four to six hours before coming to the hospital — and if you live in a rural area and your local hospital doesn’t have a PET scanner, good luck to you. When you go to the hospital, you will be injected with radioactive material, then you must wait an hour while it washes out of your body. After that, you enter the PET scanner and try to stay still for 30 minutes while the radiologists acquire the image. After that, you need to physically stay away from the elderly, young and pregnant women for up to 12 hours because you are literally semi-radioactive.
Another obstacle? PET scanners are concentrated in major cities because their radioactive tracers must be produced in nearby cyclotrons – compact nuclear machines – and must be used within hours, limiting access to rural and regional hospitals.
But what if you could use AI to convert CT scans into PET scans, which are much more accessible and affordable? That’s the pitch RADiCAITAn Oxford spinout that this month exited Theft in $1.7 million in pre-seed funding. Boston-based startup, which is a top 20 finalist Startup battlefield A TechCrunch Disrupt 2025Just raised $5 million to advance its clinical trials.
“What we’ve actually done is we’ve taken the most restrictive, complex, and expensive medical imaging solution in radiology, and we’ve replaced it with the most accessible, simple, and affordable one, which is CT,” Sean Walsh, CEO of RADiCAIT, told TechCrunch.
RADiCAIT’s secret sauce is its foundational model — a generative deep neural network invented in 2021 at the University of Oxford by a team led by Regent Lee, the startup’s co-founder and Chief Medical Information Officer.
The model compares CT and PET scans, mapping them and picking out patterns of how they relate to each other. Sina Shahandeh, chief technologist at RADiCAIT, describes anatomical structures as connecting “distinct physical phenomena” by translating them into physiological functions. The model is then instructed to pay extra attention to certain features or aspects of the scan, such as certain types of tissue or abnormalities. This focused learning is repeated many times with different examples, so the model can identify which patterns are clinically important.
TechCrunch event
San Francisco
|
October 27-29, 2025
The final image that goes to doctors for review is created by combining multiple models working together. Shahandeh compared the approach to Google DeepMind’s AlphaFold, the AI that has revolutionized protein structure prediction: Both systems learn to translate one type of biological information into another.
Walsh claimed that the team at RADiCAIT could mathematically prove that their synthetic or generated PET images were statistically similar to real chemical PET scans.
“That’s what our trials show,” he said, “that the same quality decisions are made when a doctor, radiologist or oncologist gives chemical PET or [our AI-generated PET]”
RADiCAIT does not promise to replace the need for PET scans in certain therapeutic settings, such as radioligand therapy, which kills cancer cells. But for diagnostic, staging and monitoring purposes, RADiCAIT’s technology could make PET scans obsolete.
“Because it’s such a limited system, there’s not enough supply to meet the demand for diagnostics and theragontics,” Walsh said, referring to a medical approach that combines diagnostic imaging (eg, PET scans) with targeted therapies to treat the disease (eg, cancer). “So what we’re trying to do is pick up the slack on the diagnostic side. The PET scanners themselves should pick up the slack on the theragnostic side.”
RADiCAIT has already begun clinical pilots specifically for lung cancer testing with large health systems such as Mass General Brigham and UCSF Health. The startup is now pursuing an FDA clinical trial — a more expensive and involved process than RADiCAIT’s $5 million seed round. Once this is approved, the next step will be to conduct a commercial pilot and demonstrate the commercial viability of the product. RADiCAIT intends to run the same process — clinical pilot, clinical trial, commercial pilot — for colorectal and lymphoma uses.
RADiCAIT’s approach to using AI to provide valid insights without the burden of difficult and expensive testing is “widely applicable,” Shahandeh said.
“We are exploring extensions across radiology,” added Shahandeh. “Expect to see similar innovations linking the domains of materials science to biology, chemistry and physics wherever nature’s hidden relationships can be learned.”
If you want to hear more about RADiCAIT Join us at Disrupt October 27 to 29 in San Francisco. Learn more here.
