Yeah, yeah.

And then in 2017, 2018, the Transformers pa- paper came out of Google, where sort of the, the language understanding also became much better.

And, um, so for me, I mean, personally, that 2012 paper, uh, was a big moment.

I mean, we were like, "Okay, I mean, this, there is a lot of image understanding." In my previous startup that I was working on, I was basically doing advertising technology.

So showing people the right ads based on what they have browsed for, what they have recently seen, right? And in that, one of the things that we were exploring was that if you are looking at a picture of, let's say, Deepika Padukone, and she's wearing a dress, can I sort of find similar dresses on Amazon, on Flipkart, on other, other Myntra, Ajibong, and so on, and allow that person to click on that and buy, buy that dress or similar dress right then and there? And that required image recognition, because you're now trying to map that particular picture to something that those stores are shop- are, are selling, right? So, uh, I was already working on some image recognition algorithms.

And then I was like, I mean, this technology is very good, that it can understand images really well.

And, uh, my hypothesis was that can we apply this technology to radiology imaging? And if we were to give it, let's say, millions of images, radiology images, would it be able to understand that well? And would we be able to automatically report X-rays and CT scans using the same image recognition neural networks that, uh, were used for understanding regular, regular everyday images? And so that was the...

Yeah

... from the doctor's point of view.

So what happens is patient comes in, uh, takes an X-ray.

X-ray is automatically uploaded to our system, right? Or patient comes in, takes a CT, CT's uploaded to our system.

Um, we process it immediately.

It takes, um, anywhere from a few seconds to maybe a minute to process those scans.

Um, and the results are then available as a small widget on the radiologist's workflow.

So as a radiologist reading a scan, there will be a small queue at the bottom of the screen where they can, uh, see the results of what Qure has, um, uh, Qure has p-processed for that particular scan, what the abnormalities that Qure has identified.

Yeah

Can you give us some kind of a picture, Prashant, on how exactly you see the future panning in terms of your quick traction from where you are right now? I

think India...

the biggest healthcare provider is the government.

And if you look at tuberculosis, which is the first sort of product that we launched and the first problem that we solved, more than one-fourth of the world's TB cases come from India.

So India is by far the largest, highest burdened TB country.

So we really believe that there is a substantial opportunity for us to make an impact on tuberculosis in India.

And one of the projects that we did in Mumbai is a very interesting one where we said that every x-ray taken in the BMC will automatically be processed by cure.

Take us through those days.