When Charles Onu’ cousin was both born with birth asphyxia and later developed a hearing condition, the seed for Ubenwa, an audio biometric company geared at identifying neurological disorders in infants, was also planted.
Onu, CEO and co-founder, said he hopes Ubenwa will change the way physicians provide care to infants.
“We’re trying to bring the world to a point where infant cries are considered to be a vital sign just as much as we would consider their heart rate to be a vital sign,” he explained in an interview with TechCrunch.
After Onu graduated with his undergrad in engineering he volunteered with health-based NGO’s and saw more cases like his cousins — though some were far grimmer.
The Montreal-based company uses an infant’s cry to determine if it falls within the range of having signs of a neurological condition. Crying audio is collected at partner hospitals, internationally, and then categorized into what is considered normal or abnormal . From there, the company can predict whether a child may be suffering from a potential disorder.
For now, the AI-powered software only identifies early signs of birth asphyxia, and can potentially determine learning milestones based on cry triggers. According to Onu, the company is hoping to evolve the technology to identify congenital heart disease.
It should be noted that a recent Stanford University study identified that some FDA approved AI medical devices are “not adequately” evaluated and there are no best practices set for the development of these technologies.
But to Ubenwa’s credit, they are performing studies in pursuit of building out the case for the science behind their tech further still.
For those actually using the application, they record the baby crying and then receive weekly summaries of patterns. If there is an abnormality detected, the application notifies the user and provides data to share with doctors.
“Today, doctors use physical assessment to look at eyelids, look at the skin tone, and so on and so forth,” Onu said. “If [doctors] are really worried it could be with an MRI or a brain MRI machine because that’s the ultimate standard, but we don’t live in an MRI machine every day. That is costly. With simple cry analysis you can track neurological biomarkers on an ongoing basis, non invasively.”
Despite the company being based out of Canada, they have additional operations in Nigeria (where Onu is originally from) and Brazil. The company partners with hospitals in those regions in order to get larger sample sizes for their data. Despite the fact that users will be recording cries, Onu said, those cries will not be collected and store: analysis of the cries will be matched to their existing database.
Although Ubenwa has a focus on an infant’s cry, other companies are using audio biometrics to help diagnose other conditions, but typically for an older age group. StethoMe says they use breath to identify air pathway disorders in children and share that data with physicians. Similarly, Ellipsis Health claims to use voice biomarkers to diagnose depression in patients.
To date, Ubenwa has raised $2.5 million USD ($3.2 million Canadian) in a pre-seed round led by Radical Ventures with participation from AIX Ventures, Yoshua Bengio, a fund co-founded by AI researchers and entrepreneurs Pieter Abbeel, Richard Socher, Google Brain’s Hugo Larochelle and Marc Bellemare.
Up to now the company’s focus has been on developing the technology, but this round’s funds will allow the company to do a beta launch and begin trials. Additionally, it will allow the company to potentially begin integrating their technology into baby monitors, according to Ubenwa.
Once trials are complete, Onu hopes to apply for FDA and Health Canada approvals.
“There’s a lot of guesswork doctors have to do with babies,” Onu said. “But a lot of (infant care) is really making informed guesses as to when to make this action or where to make that action, and we’re hoping to close some of these gaps.”
source https://techcrunch.com/2022/07/25/can-a-babys-cry-identify-neurological-disorders-ubenwa-says-yes/
0 comments:
Post a Comment