A startup is exploring how the artificial intelligence (AI) technologies used in self-driving cars could act as a “blueprint” for smarter surgeries.
Theator, based in Palo Alto, Calif., is the developer of a “surgical intelligence” platform designed to improve the decision-making of surgeons and, therefore, potentially boost patient outcomes.
The startup’s platform provides lists of steps during procedures, visual content at “crossroads” points for medical professionals to better decide which direction to take, milestone recommendations to adhere to best practices, and more.
However, Theator wants to go further and utilize AI in its quest to “become the brain behind autonomous surgery.”
The way to do this? Study what’s happening in the autonomous vehicle space.
Theator is a member of NVIDIA Inception, an AI incubator. According to Theator CTO Dotan Asselmann, the company has started examining how self-driving car models can contribute to a roadmap for safe, AI-assisted surgery.
In a similar fashion to how self-driving car AI models analyze video content and maps to correct and improve driving on the roads, a similar AI algorithm could be trained to improve the surgical review process, both during and post-procedure.
“Just as autonomous carmakers want to stop a vehicle before an accident, Theator wants to stop surgeries before any mistakes,” Nvidia says. “And it’s doing this by analyzing video taken of surgeries being performed all over the world.”
Surgical interns will often learn their craft from experienced professionals in the field and by performing rotations in different specializations, and potentially at a number of hospitals. However, Theator’s scalable AI aims to make their education global by analyzing thousands of surgeries to standardize the process.
“Horizontal data sharing between surgeons has been limited — it’s mainly happened at conferences,” Asselmann said. “In today’s COVID reality, surgeons’ ability to expand their knowledge at scale is stifled.”
Surgeries are sometimes caught on camera, but the system Theator is working on is based on an edge appliance secured to a laparoscopic cart. Nvidia’s Jetson AGX Xavier platform captures and processes video content, where Theator’s software then strips out any identifying marks for anonymization and uploads it to the cloud.
The AI model then generates ‘highlight’ packages, created through the analysis of a wider surgical video pool, which focuses on “select critical minutes where important decisions were made.”
This allows the platform to separate surgical video data into steps, events, decision-making points, and milestones for review.
It is hoped that in the future, the model could help predict where inventions may be necessary, reduce mistakes, and improve surgical standards as a whole.
“Surgeons are inundated with endless parameters flowing from multiple directions during surgery,” Asselmann commented. “Our objective is to reduce the cognitive overload and aid them in making the optimal decision at the right time, for the right patient and circumstances.”
Previous and related coverage
Have a tip? Get in touch securely via WhatsApp | Signal at +447713 025 499, or over at Keybase: charlie0
