Well, here’s a new one for you—an AI program that loves crappy healthcare software.
When Sean Lane, a former NSA operative who served five tours of duty in Afghanistan and Iraq, first entered into the healthcare-AI arena, he was overwhelmed with data silos, systems that don’t speak to each other, and many, many portals and screens.
Instead, Lane and a team taught an AI system to use software that already exists in healthcare just like a human would use it. They named it Olive.
“Olive loves all that crappy software that healthcare already has,” said Lane. “Olive can look at any software program, any application for the first time she’s ever seen it, and understand how to use it.”
For example, Olive navigates electronic medical records, logs into hospital portals, creates reports, files insurance claims, and more.
Olive does so thanks to three key traits. First, using computer vision and Robotic Process Automation, or RPA, the program can interact with any software interface just as a human would, opening browsers and typing. Second, machine learning enables Olive to make decisions the way human healthcare workers do. The team trained Olive with historical data of how healthcare workers perform digital tasks, such as how to file an insurance eligibility check for a patient seeking to undergo a procedure.
Finally, Olive relies on a unique skill that Lane developed based on his work at the NSA identifying criminals across disparate government sources—the ability to match identities across databases. Just as NSA software can determine if a terrorist in the CIA database is the same as in the Homeland Security database, so Olive matches a patient across disparate databases and software, such as multiple electronic healthcare record programs.
Lane formed a healthcare tech company in 2012 and began building the Olive software in 2017. Today, Olive—both the company and the software name—is 120 employees strong in Columbus, Ohio. The software became commercially available 18 months ago and is currently being used in 195 hospitals and healthcare companies, Lane told IEEE Spectrum.
So far, those users have primarily put Olive to work automating insurance eligibility checks and claim status denials, said Lane. In the future, he hopes users will also adopt the tech to handle medical records and identify cohorts for clinical trials.
Like Steve Jobs and his fruit-named company, Lane has grand visions for his technology. Once Olive bots are widely adopted, Lane surmises, they could be used to communicate among healthcare organizations and to be queried like a Google for healthcare. This “Internet of Healthcare,” as Lane calls it, would connect previously siloed healthcare systems into a convenient, searchable online community enabled by AI.
But for that to happen, Olive bots would need to be widespread among healthcare computers. Like the early days of Arpanet, “we need everyone to buy a modem, and that modem is Olive,” Lane told Spectrum.
Another challenge to an Internet of Healthcare is privacy. Rules for sharing patient data can be built into each bot at a hospital or healthcare center so that they share data only for specific business or medical reasons. “It’s very difficult, but it is doable,” says Lane. Similar systems are in place in intelligence communities, he adds.