Digital twins take off in health care

If you haven't heard of digital twin technology in health care yet, you probably will soon.

Why it matters: Virtual likenesses of a human or an object are increasingly being seen as a powerful tool for research, writes Axios' Megan Morrone.

Zoom out: Right now "digital twin" has become shorthand for an array of AI models and can mean anything from a statistical model of a complex organism, to a video avatar of a billionaire trained on his speeches and posts, she writes.

• In health care, digital twins can be an AI model of a person or a part of a person used to analyze large quantities of health data in order to provide more personalized treatment or speed up drug development.
• They're not entirely new. Studies involving digital twins first began popping up in PubMed in 2017.
• Digital twins offer "tremendous opportunities for personalized healthcare," but they also pose potential ethical, societal and legal challenges, according to a review article in npj Digital Medicine.

Tina's thought bubble: I've struggled to fully wrap my mind around this concept. Lisa Shah, chief medical officer of a startup called Twin Health, told me to think about it like the challenge of building autonomous cars.

• "There's lots of inputs and outputs, lots of things that can go wrong when you're driving a car, so you have to almost create a digital replica of that car and be able to predict what's going to happen," Shah said.
• "We do the same thing," she said, only it's with digital twins of patients using Bluetooth-connected sensors to measure data points like blood pressure and heart rate.