Meet BioticsAI, a startup that has built an AI-based platform that plugs into an ultrasound machine to prevent fetal malformation misdiagnosis. BioticsAI has been quietly working on its solution for the past two years and is now unveiling its product onstage at TechCrunch Disrupt as part of the Startup Battlefield.
“BioticsAI processes all of the images and videos captured during the screening to localize fetus malformations, validate the quality and completeness of the screening and then extract all the information to automatically generate reports,” co-founder and CEO Robhy Bustami told me before the TechCrunch event. Bustami has some personal knowledge of the field, as his mother is an OB/GYN. He co-founded the startup with Salman Khan, Chaskin Saroff and Dr. Hisham Elgammal.
But first, let’s talk about the problem BioticsAI is trying to solve. In many countries or rural areas, there’s a shortage of obstetrics professionals. While the World Health Organization has designed a thorough process for prenatal ultrasound scans, it can be easy to miss a fetal malformation if you’re short on time and staff.
BioticsAI is all about making sure that ultrasound screenings are conducted properly in an efficient manner. By speeding up some parts of the process using artificial intelligence and automated report generation, doctors can also save time and reduce their overall workload.
There are three different parts that make BioticsAI work. First, the platform can be used to identify when screenings aren’t being done completely. The software solution integrates with a wide range of picture archiving and communication systems (PACS) and medical imaging devices so you can just add it to your existing workflow without any expensive hardware investment.
“Our goal isn't to build like the next $200,000 ultrasound machine. Our goal is to be able to integrate with an ultrasound machine that's been around for eight years and significantly improve the quality of that screening,” Bustami said.
After that, BioticsAI looks at images and see if images of the fetus have been captured properly from all angles according to international standards. Mistakes happen, but it’s important that doctors can learn from them and improve so that the same mistake doesn’t happen again and again.
As an OB/GYN, once you’ve mitigated the “operator-dependent problems,” as Bustami calls those problems, BioticsAI helps you reach a diagnosis more efficiently. This is where the artificial intelligence part of the product kicks in. The company analyzes the images to help doctors identify a fetal malformation. Right now, the company claims that BioticsAI has reached a 96% accuracy.
The startup has already built a comprehensive dataset of over one million prenatal ultrasound images and is always looking at new partnership opportunities to grow its database and improve the accuracy of its algorithm. Of course, BioticsAI isn’t sending conclusions to patients directly. Instead, it can “bring that abnormality to the attention of the interpreting physician so that it doesn’t go unnoticed,” Bustami said.
Finally, the platform extracts relevant data from images and automates report generation. This is all about saving time.
“On average, OB/GYNs are spending 15 minutes per patient on generating reports and documentation. And our goal with our product is to just mitigate that by automatically generating these reports and extracting all this information, providing valuable time savings back to the physician and back to the hospital,” Bustami said.
BioticsAI has raised a pre-seed round from Techstars Boston and Blackbird Venture's pre-seed program that helps when it comes to commercialization (Startmate). It has started several pilot programs and clinical studies with eight institutions.
There are other companies working on ultrasound analysis, such as Ultrasound AI, Origin Health and Sonio AI. They’re all tackling this problem from a different angle with a specific niche focus.
“We focus the majority of our AI efforts on second trimester anomaly screenings because this is where 90% of potential fetal abnormalities are screened for,” Bustami said. But if BioticsAI works well, the company could soon expand to adjacent fields, such as gynecology, urology and neonatology. It could become the AI diagnostics platform for reproductive health and beyond.