ATS Breathe Easy - ATS 2026: How Data Science and Precision Medicine go Hand-in-Hand

So you both kind of touched on this concept of precision medicine and the better use of data as we t- look at our critical care population, and that kind of connects the larger theme of the conference with AI, data science, big data.

How is your own work and the work of other mentors and collaborators and mentees really contributing to that knowledge?

Well, I think, um, i- in two spaces, you know.

So one is trying to get the questions a bit more specific and clinically actionable, and I think we're really getting there.

And so we've got these, uh, really interesting ideas of syndromes that have moved to more like subphenotypes and thinking about whether those respond differently to key treatments.

The other end of things is really the implementation science around this.

So we have these great models.

We have some early evidence that they might actually respond differently like we suggested, but what do you do with that? How does a team make use of that information? When do they need it? How do you bring that to them? And, and so seeing the science around that is really exciting.

That's great.

I think one big change that we've seen t- over the past few years really is that there's more and more infrastructure to do this collaboratively and at scale that didn't exist when, you know, Pat and I were starting our research careers.

Um, so Pat and I are both members of the Cliff Consortium, which is a multicenter, uh, effort to standardize ICU data so that it can be used for studies acr- uh, across centers for, uh, whether it's sort of in-depth analyses of ventilator physiology or delivery science.

And, you know, that required a collaborative effort of multiple sites to bring-- to, to come together and agree on what data elements actually need to be included in these studies, but it's now enabling team science across the country and actually really centers even outside the US.

And on the biological scale, it's-- on the biological side, it's possible to do biology on scale, multiple modality work.

Uh, there's-- Our new lenses in our complex patients that we didn't have available even a few years ago, and we're starting to see the fruits of that work.

I love how you're both talking about crossing silos, if you will, beyond institutions, beyond different types of research, the biological sciences, the data science, the implementation science.

So you both kind of touched on this concept of precision medicine and the better use of data as we t- look at our critical care population, and that kind of connects the larger theme of the conference with AI, data science, big data.

How is your own work and the work of other mentors and collaborators and mentees really contributing to that knowledge?

Well, I think, um, i- in two spaces, you know.

So one is trying to get the questions a bit more specific and clinically actionable, and I think we're really getting there.

And so we've got these, uh, really interesting ideas of syndromes that have moved to more like subphenotypes and thinking about whether those respond differently to key treatments.

The other end of things is really the implementation science around this.

So we have these great models.

We have some early evidence that they might actually respond differently like we suggested, but what do you do with that? How does a team make use of that information? When do they need it? How do you bring that to them? And, and so seeing the science around that is really exciting.

That's great.

I think one big change that we've seen t- over the past few years really is that there's more and more infrastructure to do this collaboratively and at scale that didn't exist when, you know, Pat and I were starting our research careers.

Um, so Pat and I are both members of the Cliff Consortium, which is a multicenter, uh, effort to standardize ICU data so that it can be used for studies acr- uh, across centers for, uh, whether it's sort of in-depth analyses of ventilator physiology or delivery science.

And, you know, that required a collaborative effort of multiple sites to bring-- to, to come together and agree on what data elements actually need to be included in these studies, but it's now enabling team science across the country and actually really centers even outside the US.

And on the biological scale, it's-- on the biological side, it's possible to do biology on scale, multiple modality work.

Uh, there's-- Our new lenses in our complex patients that we didn't have available even a few years ago, and we're starting to see the fruits of that work.

I love how you're both talking about crossing silos, if you will, beyond institutions, beyond different types of research, the biological sciences, the data science, the implementation science.