CEO says poor healthcare data infrastructure is hindering AI innovation in the field
It is undeniable that the hype of AI in healthcare is undeniable. AI startups are dominating the digital health investment market, with companies like Abridge and Ampience Healthcare surpassing unicorn status, and the White House recently released an action plan to drive the use of AI in key areas such as healthcare.
But, one digital health executive believes that unless a key issue is addressed, AI in the field may soon stagnate.
According to Mitesh Rao, CEO of Omny Health, a national data ecosystem that promotes medical research, the problem is the data infrastructure. Rao believes that scalable AI in healthcare depends on access to extensive, representative and interoperable data.
He said in an interview last week that the current system is decentralized and inaccessible, making meaningful AI development difficult.
These systems are isolated because information is stored on multiple platforms that lack standardized formats and shared data exchange channels. Additionally, current suppliers have little incentive to promote better data sharing.
Rao noted that CMS launched a new interoperability plan last week – while such efforts are good, they wouldn't have much impact if they don't provide obvious incentives to incumbents like Epic or Cerner.
This all creates a patchwork of locked data infrastructure, which makes it challenging for developers to access the data they need to create advanced AI solutions.
“Most of the successful AI efforts in healthcare today focus on data that are not difficult to access, such as documentation or revenue cycle management, are not necessarily related to proprietary and deep patient data,” Rao said.
He noted that data limitations are a barrier that innovators are difficult to overcome when they begin to explore more complex AI use cases in healthcare, especially when it comes to applications that touch clinically.
Overall, Rao believes that medical technology leaders need to “build roads in front of Ferrari.” In other words, an ambitious AI project requires a basic infrastructure first.
Image: filograph, Getty Image
