Deferring to your considerable expertise and experience here, can you tell us a bit more about the problems with cancer diagnosis?

What do we need to do better? So actually, we focused on prostate cancer first because it's one of the most common and difficult cancers to diagnose.

In the past, we used to screen for prostate cancer using the PSA blood test and a thing called the digital rectal examination, which I won't I don't need to go into more detail about it.

If those were positive, you would have a biopsy.

The result was that men with even the lowest grade group one prostate cancer ended up being over-treated in very large numbers by removing or irradiating the gland.

The side effect rates were horrendous.

Now, it's actually since been shown that that grade group 1 low grade prostate cancer doesn't metastasize.

So how does your AI solve these problems in technical terms? What data inputs does it need and where can it be used in the diagnostic pathway?