Pages Navigation Menu

SHOWFUN - Show & Fun & More!

AI doctors will beme 'as ubiquitous as stethospes'

From helping humans live longer and hacking our performance, to repairing the body and understanding the brain, WIRED Health will hear from the innovators transforming this critical sector. Read all of our WIRED Health coverage here.

One of the biggest problems facing doctors isn't patients' injuries or illnesses – it's the sheer quantity data. Most will spend more time going over medical records than actually dealing with their patients.

It's a problem that "AI doctors" could help address, with supercomputers processing information far faster and more efficiently. The problem, IBM's Kyu Rhee tells the crowd at WIRED Health, is trust.

"Studies have shown that if a doctor wears a stethoscope, you trust him or her more. But in 1816, Dr René Laennec, a French physician, was examining his patient, trying to listen to her heart sounds with his ears," said Rhee. "He took 40 pieces of paper, rolled it up, and created the first stethoscope."

Looking to the future, Rhee sees a "cognitive system" such as IBM's Watson supercomputer having a similar role to play in human healthcare. Such systems, he said, will become as ubiquitous as the humble stethoscope.

Rhee, who was a physician earlier in his career, recalls struggling with the sheer volume of data involved in treating patients. Worse, the data was presented, at the time, on reams of paper and charts. Throw in new materials and understanding generated by medical journals and it soon becomes a mountain of information that can hinder, rather than help.

"It's humanly impossible for me to know everything that's in every one of those charts," said Rhee. "As a primary care physician, I had maybe 2,000 patients that I took care of. To know everything about every person was very, very challenging. I would typically get 15 minutes with each patient [...] half of that time was spent documenting them."

Using systems such as Watson – perhaps best known for winning the US gameshow Jeopardy! in 2010 – all that data can be translated at the point of care into usable, relevant data. It also allows doctors to share information and expertise, potentially on a global scale.

Democratised data could also help improve doctor/patient relations, and improve pre-emptive diagnoses.

"Each of us, as individuals, will produce in our lifetimes 300 million books worth of data," he explained. "It's the data that's on your wearable; it's the data that, as genomic testing becomes more prevalent, is related to that; it's the data in terms of the social and financial determinants of health; and it's the healthcare data in your medical records. There's an opportunity now to translate it."

Rhee suggests future wearables could, with the user's permission, log activity and share it with a family doctor, alerting them when there are sharp declines that could indicate physical or mental health dips. It wouldn't just be a simple pedometer either – blood sugar, pressure and sleep patterns could also be sent to doctors, pre-empting potential problems.

"Why are we in a system that reacts, more than being proactive?" he asked. "[One that] can predict, that can personalise, that can prevent these health outcomes that are frankly bankrupting many of our economies.

The role of an AI doctor isn't intended to replace the human component, but rather complement it. "It's about augmenting or complementing; the role of the human doctor is in building the relationship. It's the skills that humans have – the compassion, the ability to abstract, to generalise, to have common sense, to have morals. These are all key aspects. Technology should be used to improve these relationships, not make them harder."

Leave a Comment

Captcha image