Can a supercomputer with artificial intelligence take better care of you than your doctor can?
In addition to partnering with Celgene (celg) to better track negative drug side effects, IBM (ibm) is applying its cognitive computing AI technology to recommend cancer treatment in rural areas in the U.S., India, and China, where there is a dearth of oncologists, said Deborah DiSanzo, general manager for IBM Watson Health.
For example, IBM Watson could read a patient’s electronic medical record, analyze imagery of the cancer, and even look at gene sequencing of the tumor to figure out the optimal treatment plan for a particular person, she said.
“That is the promise of AI—not that we are going to replace people, not that we’re going to replace doctors, but that we really augment the intelligence and help,” DiSanzo said.
Athenahealth CEO Jonathan Bush, however, disagreed. “It’s OK—we’re friends here—of course you’re going to replace me!” he exclaimed, noting the limitations of traditional doctors.
“The human is wrong so freaking often, it’s a massacre,” said the co-founder of Athenahealth, which sells cloud-based electronic health records software to hospitals and doctors’ offices. “Nobody ever goes after the radiologist—they’re wrong so often we don’t blame ’em.”
While less enthusiastic about artificial intelligence’s current contributions to healthcare, Bush suggested a perhaps more radical vision of the future, in which machines do indeed supplant many rudimentary medical functions. “Forget brain scans and cancer silhouettes—what about plain film and broken bones? Wouldn’t Watson be a lot more reliable with that?” he asked rhetorically. “And isn’t that billions and billions of dollars of some radiologist looking at it and saying, ‘Yup, that’s a broken bone?'”
Rather than replace the radiologist altogether, DiSanzo said Watson would make the doctor’s job easier by narrowing down a series of X-ray or CT images to the most relevant scans, and combining that with other patient records to “make your definitive diagnosis quicker, faster.”
Meanwhile, even 3D printing can’t replace doctors and old-fashioned medical technology yet, according to Dr. Anthony Atala, director of the Wake Forest Institute for Regenerative Medicine, who also participated on the conference panel.
With the shortage of donor organs becoming an increasingly acute crisis, Atala has been able to grow and transplant skin, cartilage, muscle, bladders, vaginal organs, kidney cells and more, engineered using patients’ own tissue and organs.
But there’s a big caveat. “All these were created by hand,” Atala said, from the initial biopsy to the manufacturing to the transplant itself. “So the concept really is, how can we automate it, how can we scale it up? And that’s where 3D printing comes in.”
Bush also hinted that AI technology such as IBM’s Watson might be used in conjunction with Athenahealth’s technology, which could electronically survey patients in order to collect data points that might be early warning signs for disease. “The real opportunity for artificial intelligence,” he said, is detecting signs of disease sooner in the many people who “go from stage one to stage two to stage three cancer” before it’s caught by a doctor. “I’m one of them! I’m one of the problem people,” he added. He later told Fortune that by “I,” he meant “a theoretical patient who is being identified by their data signature.”
Already, Athenahealth is beginning to analyze its data in order to try to identify the onset Alzheimer’s disease before it has been diagnosed. “The idea is that [patients’] regular phenotypical signature, complemented by AI and some incremental outreach or surveys, could speak for them,” Bush said. It “could raise their digital hand before the disease progresses too far.”
Editor’s note: An earlier version of this story suggested that Bush’s comment “I’m one of the problem people,” might refer to an ailment he had personally that had gone undiagnosed. The article has been updated to reflect a response from Bush clarifying his earlier comments.