Can you walk us through how you designed that study and why this data set was particularly suited for this question and how you selected your cohort?

Yeah, so this data set, the All of Us Research Program, was created by the National Eye Institute, and one of the advantages of this data set is that you have access to patients' electronic medical records, although they're de-identified.

But getting access to the electronic medical record really was essential for this study because the pooled cohort equation that was developed by the American Heart Association uses data that is in the electronic medical record and usually collected.

So with the All of Us Research Program, you get access to that data in terms of availability in the electronic medical record.

You also get access to the ICD 9 codes and the CPT codes, so you can look at first diagnosis of AMD that's coded, for example.

In terms of the study design, the desire was to have the ability to really use this PCE score to help predict the risk of an eye disease.

So you wanna have follow-up And so the study was designed to end around 2015, 2009 to 2015, so that then there could be at least five to six, seven, eight years of follow-up.

As far as future directions, if you had to put, like, your crystal ball there, how do you envision that we are going to incorporate this, both at our end, um, versus how primary care physicians might incorporate a score related to an eye finding? How do you imagine this is going to unfold, um, and how are we going to harness the power of artificial intelligence and bring this actually to patient care?