AI, machine learning

IBM CEO Ginni Rometty gave a powerful keynote address at the annual HIMSS meeting in Orlando that kicked off Monday describing humanity as on the cusp of ushering in a new cognitive era.

She was careful to characterize artificial intelligence in healthcare as merely augmenting the intelligence of doctors and not replacing it. Many will be rightly skeptical — radiology is a field that is bound to feel some pain as machines learn to read medical images quickly and more comprehensively than humans — but for now let’s highlight the clear benefits.

Specifically, the ones that Mayo Clinic is experiencing.

At a panel discussion on the impact of AI in healthcare, the venerable institution’s chief information officer and a breast cancer oncologist were on hand to describe AI in action. For instance, how efficiently IBM Watson Health can comb through a patient’s medical record — be it structured data like lab tests and imaging as well as unstructured data like doctor’s notes — to match a cancer patient to a clinical trial.

It’s widely known that patient recruitment is cancer trials is slow.

Part of the challenge in low participation is the multitude of clinical trials that are currently available with each of them having lengthy inclusion/exclusion criteria.

“It’s also written in technical English language. It’s not structured. It’s not consistent from trial to trial,” explained Cris Ross, CIO of Mayo Clinic in Rochester, which has been doing a trial with IBM Watson Health’s Clinical Trial Matching.

Enter IBM Watson Health, the data cruncher.

“Watson understand the inclusion/exclusion criteria. It has read all of ClinicalTrials.Gov. It has read all of our clinical trials. It has read all of the world’s literature on oncology and then it applies against a subset of patients records to provide recommendations.”

Consider this power of AI against the exponential increase in human scientific knowledge. Specifically, and very narrowly, look at the big jump in new drugs approved in the last five years.

“Even as a sub-specialist it’s hard to keep up with this knowledge rate,” explained Tufia Haddad, a breast cancer oncologist at Mayo Clinic, noting that Mayo has 57 breast cancer clinical trials that are enrolling patients. “I can’t possibly remember all 57 trials and the eligibility criteria for each one of those and [then] put it into the patient context.”

Now with Clinical Trial Matching there has been a significant improvement in just this process.

“30 minutes is down to 8 minutes for one of our research coordinators to screen a patient for a clinical trial,” Haddad declared. “That coordinator can now spend that valuable time gained … in educating the patient on why it’s important for her to be in that clinical trial, helping to break down other barriers.”

While Haddad was making the case for how AI is changing the way she and her team practice cancer care even for prostate cancer, another panelist highlighted how AI adoption is the tool t actually make such changes mainstream.

Chris Belmont, CIO of the University of Texas MD Anderson Cancer Center talked about how in the quest to make cancer history — the stated mission of this renowned organization — it’s imperative to move AI from research capacity to mainstream medical practice.

Full story and credit here:

Mayo Clinic CIO on AI: This stuff is really real