“It's really it is a combination of a whole range of different techniques that we're using to come up with the best possible selection model,” she adds.
Once Life Whisperer’s models have been validated – an ongoing process that is “looking really good” Perugini says – the aim is to offer the technology as a ‘drag and drop’ cloud-based application to clinics.
"They'll get an instant report using a validated, evidence-based model and that will give them basically an assessment of which embryos they should implant and why. It's a scalable product that can be put in any clinic in the world,” says Perugini.
"What's really important is that it doesn't impose any significant process change for them. They're already taking images. They were doing that manual assessment under the microscope. All they need to do is drag and drop those images."
The use of AI to analyse medical imagery has huge potential and has been the focus of numerous studies. In June researchers at University of Adelaide applied deep-learning image analysis techniques to CT scans of major organs and tissue in patient chests to determine their lifespans.
The platform underpinning Life Whisperer could be used for other medical diagnoses that are based on imagery, Perugini said.
“We built a general infrastructure that can actually be used to build accurate, valid models for other medical applications leveraging the same base technology and bringing together the same types of techniques but for a different medical diagnosis application. So absolutely we think we can leverage this into many different use cases within the medical diagnostic area. And that's certainly our intention beyond fertility,” she explained.
The company is now actively seeking Series A investment.
Sign up for Computerworld eNewsletters.