Short answer: Sam Nadler demos a digital twin dashboard showing manual selection over-indexing on tier one academic centers, with slow enrollment and missed diversity targets, while AI-optimized selection recovers a two month delay and hits diversity targets.
Watch this answer from Episode 34 (starts at 3:13) · Episode 34 notes and transcript
From Episode 34: AI Is Rebuilding Clinical Trials, starting at 3:13.
Sam Nadler [3:12]: Hit that button and be reminded on every Friday. And the docket, we're gonna demo a little something we built with in mind. And, like we do in every episode, this may be totally off base. I did my best to try and kind of capture maybe something that would be interesting for Ryght. But as you can see, this is a digital twin dashboard and could show the difference between manually selecting a site versus using AI. So let me kind of walk you through this trial strategy snapshot, and let's start a tour, actually. I think that would be really helpful. So, first, the digital twin. This dashboard simulates the clinical trial network using historical performance data and predicts enrollment velocity, diversity, and risk in real time. Obviously, these trials, the goal is to, I'm assuming, finish on time and either under or at cost and with the right markers involved. So the status quo problem, we're currently simulating a standard manual site selection strategy. Most sponsors over index on tier one academic centers where enrollment could be slow and diversity could be below target. With that, there could be financial consequence. So inefficiency has a price. You could run over time.
Sam Nadler [4:40]: You could run over in terms of cost. There could be regulatory risk, and with that, we can have AI optimized output. So the strategic advantage is, we can recover the two month delay and hit diversity targets. So, with this AI optimized output in terms of site selection, maybe, Emory Healthcare could be a great option versus, a tier one clinic that's ranked a little bit lower like the Mayo Clinic. So is this somehow you know? And you could, like, trigger these simulation parameters that have, biomarker complexity. How does that change the ranking of these sites, which could be a better fit for the clinical trial? Overlay the census overlay. You have the potential strategic tension with equity. Is is this somewhere in the realm of, like, how your business is thinking about site selection and why it matters?
