Femtech company becomes first foetal ultrasound start-up using AI to get FDA clearance

The French femtech start-up Sonio has announced its AI assistant for prenatal ultrasounds has been cleared by the US FDA.

The company has received regulatory class II FDA clearance for Sonio Detect, a product that uses AI to improve foetal ultrasounds and support practitioners.

The platform, its founders argue, provides an AI software that associates the knowledge of prenatal medicine practitioners with all types of medical, imaging, genetic and environmental data to “optimise” prenatal screening and diagnosis.

“Bringing the first easy-to-use, manufacturer agnostic (GE, Samsung and Canon) and efficient quality control solution to all OB-GYNs, maternal-foetal medicine specialists and sonographers, we believe better screening will lead to better detection of potential anomalies or reassurance to provide better maternal care,” said Sonio’s CEO and co-founder, Cecile Brosset.

“Our goal is to transform prenatal care by providing a reliable tool that ensures better health outcomes for both mothers and babies.”

The start-up says its AI solution aims to enable healthcare professionals to “raise the bar” on quality control of ultrasound exams whilst ensuring improved efficiency overall.

In real-time, the product detects views, anatomical structures within supported views, and verifies quality criteria of the supported views, allowing practitioners to detect the presence of heart and brain structures and other important foetal imaging quality criteria, Sonio argues.

After announcing its most recent US$14m dollar funding round in July, the company says it is ready to expand into the US market, with the goal of spreading accessible maternal care across the globe.

Sonio co-founder, Remi Besson, said: “We are deeply committed to advancing the field of prenatal health through innovation.

“Obtaining FDA clearance for our products demonstrates our dedication to safety and efficacy in the US market. Our goal is to ensure everyone has access to top-quality prenatal care across the globe.”

According to its developer, Sonio Detect’s software was found to have a high accuracy rate in identifying correct labels, views and quality criteria, regardless of manufacturers, patients BMI, age, ethnicity, gestational age.

The tool proved its performance validating it on over 17,000 ultrasound images, with more than 92 per cent sensitivity on its ability to detect labels and types of ultrasound images for instance.