[MUSIC PLAYING]
WILLIAM FAUBION, JR., MD: My name is Bill Faubion. I'm a gastroenterologist here at Mayo Clinic.
I'm really interested in how immune cells, specifically lymphocytes in patients, develop into the types of aggressive cells that they are in patients with Crohn's disease and ulcerative colitis.
So we're learning that when an immune cell is educated to become a certain type of phenotype, we're really studying all of the epigenetic patterns that lead to that aggressive cell, and what we're finding out is we can intervene with medications that affect the epigenetic complexes themselves and change their phenotype from an aggressive phenotype to a more regulatory phenotype.
Result for the patients, I think, in the short term is that I may be able to take a sample of blood from the bloodstream, isolate a certain type of cell, let's say a regulatory cell, and then modify it in the laboratory, such that when we put it back into a patient, it can maintain its function long term, even in the setting of inflammation.
We see as many as 4,000 patients a year with inflammatory bowel disease here. That generates an awful lot of data. We can build models that can account for an unlimited number of variables, right? The math is not a problem for a mathematical model like that.
So in those ways, I think a tool, an artificial intelligence tool, can help guide us for unrecognized groups of patients, right, that we didn't recognize were a group. Then if we go back and start collecting biological variables that associate with that group, we may understand, wow, I never knew that group existed, and look, that group rarely responds to this pattern of medication. It brings solutions that may exist in the records right to the point of care, right to that clinician that's going to see the patient.
We're developing all of those data sets right now, and then we're working with the right partners to try to help us understand, how do we pull it together in a meaningful fashion that we can actually make a prediction? That's the future.
And I think the next 18 months or so, you're going to see a lot of work being done in integration of these complex data sets, and we would really like to be at the forefront of that.
[MUSIC PLAYING]
