Sep 3, 2026 · 56 min · 12 segments
What if medical scans we already collect contain signals that could predict cancer treatment response, reveal disease risk years earlier, or identify biological patterns the human eye cannot see? In…
Anant MadabhushiGuest
Doug FloraHost
Sanjay JunejaHost
Yeah, but one of the things that I heard, you know, I guess first or earliest was on the fundus.
When you're looking inside the eye, you're able to kind of get a good idea of very small vessels, and it gives you an idea of the bigger systemic picture, right? So if the stuff is happening way far away from the heart, you know, one can, you know, imagine that things are happening between the heart and to that very distant place.
What are some of those things that kind of first tipped us off like, whoa, this might be a really big thing? I mean...
There could be procedures that we're doing routinely that could potentially not be as necessary.
What did that process look like? And I believe that's where one of many things started for you when it came to this, you know, what you called the latency in data that we are bringing to the surface.
And, you know, retina fundus images are a great place to begin because every one of us who, at least most of us, hopefully go to visit our ophthalmologist.
So, you know, I've got a date with my ophthalmologist, at least on an annual basis.
And one of the things that happens, of course, is that you would get a radional scan.
The ophthalmologist is typically looking for signs of cataract or glaucoma or diabetic retinopathy or diabetic macular edema.
But the discovery that I think the community at large, because I think it's appropriate to acknowledge that there are several other groups working in the space, that the community has made has been that there's so much data here that can tell us about other systemic conditions, that the vasculature in the eye can tell us about future cardiovascular disease, can tell us about stroke, can tell us about kidney disease and kidney failure.
In fact, in our group, we're now looking at opportunities where we can predict onset of future malignancies.
We had a publication last year where we demonstrated that the vasculature in the eye could prognosticate up to 12 years out the onset of hematologic malignancies like lymphoma, leukemia, multiple myeloma.
And again, the message, I think clearly, is that if we look at this data in an opportunistic way beyond the primary reason for which it was acquired, there's a lot there.
And what I think it's also telling us is that there are subclinical changes that are being reflected in the vasculature, which if we look deeply enough, that we're going to find initial telltale signs that might essentially be a warning sign of what to expect in the future.
And this is, I think, the most exciting piece, being able to find these subtle features, but from routinely acquired data, which has the advantage that we're not going to have to generate another scan.
It already exists because it's part of ongoing examination or ongoing test.
And this has been really something that our group has been very excited in.
And then maybe if I might pivot, because I want to connect it back with something that you said earlier about vascular vessels of the tumor.
And one of the things that our group has really appreciated has been the immense amount of information that we don't really appreciate in vasculature, right? I gave you the example with retinal vasculature and being able to predict future onset of disease.
More and more I'm convinced, Sanjay and Doug, that vasculature, I think, really holds significant cues about multiple different diseases.
Yeah, but one of the things that I heard, you know, I guess first or earliest was on the fundus.
When you're looking inside the eye, you're able to kind of get a good idea of very small vessels, and it gives you an idea of the bigger systemic picture, right? So if the stuff is happening way far away from the heart, you know, one can, you know, imagine that things are happening between the heart and to that very distant place.
What are some of those things that kind of first tipped us off like, whoa, this might be a really big thing? I mean...
There could be procedures that we're doing routinely that could potentially not be as necessary.
What did that process look like? And I believe that's where one of many things started for you when it came to this, you know, what you called the latency in data that we are bringing to the surface.
And, you know, retina fundus images are a great place to begin because every one of us who, at least most of us, hopefully go to visit our ophthalmologist.
So, you know, I've got a date with my ophthalmologist, at least on an annual basis.
And one of the things that happens, of course, is that you would get a radional scan.
The ophthalmologist is typically looking for signs of cataract or glaucoma or diabetic retinopathy or diabetic macular edema.
But the discovery that I think the community at large, because I think it's appropriate to acknowledge that there are several other groups working in the space, that the community has made has been that there's so much data here that can tell us about other systemic conditions, that the vasculature in the eye can tell us about future cardiovascular disease, can tell us about stroke, can tell us about kidney disease and kidney failure.
In fact, in our group, we're now looking at opportunities where we can predict onset of future malignancies.
We had a publication last year where we demonstrated that the vasculature in the eye could prognosticate up to 12 years out the onset of hematologic malignancies like lymphoma, leukemia, multiple myeloma.
And again, the message, I think clearly, is that if we look at this data in an opportunistic way beyond the primary reason for which it was acquired, there's a lot there.
And what I think it's also telling us is that there are subclinical changes that are being reflected in the vasculature, which if we look deeply enough, that we're going to find initial telltale signs that might essentially be a warning sign of what to expect in the future.
And this is, I think, the most exciting piece, being able to find these subtle features, but from routinely acquired data, which has the advantage that we're not going to have to generate another scan.
It already exists because it's part of ongoing examination or ongoing test.
And this has been really something that our group has been very excited in.
And then maybe if I might pivot, because I want to connect it back with something that you said earlier about vascular vessels of the tumor.
And one of the things that our group has really appreciated has been the immense amount of information that we don't really appreciate in vasculature, right? I gave you the example with retinal vasculature and being able to predict future onset of disease.
More and more I'm convinced, Sanjay and Doug, that vasculature, I think, really holds significant cues about multiple different diseases.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.