So it's easy to see this tumor in the upper part of this fatty breast.

You then have some body part descriptions

Let's see you want to look at medical records and try to predict of a breast cancer as malignant or benign.

Tumors?

The problem with breast density is that it's truly the wolf in sheep's clothing.

And let's say we have a friend who tragically has a breast tumor, and let's say her breast tumor size is maybe somewhere around this value.

The cancer.

And I started talking to him about this problem with breast density, and we realized that we might be able to get this detector close enough around the breast to actually find small tumors.

And you see here in higher magnification the normal acinus and the ugly tumor.

And the CAT scan revealed no blood clots in the lung, but it showed bilateral, visible, palpable breast masses, breast tumors, that had metastasized widely all over the body.

Women suffer from that disease.

And one afternoon, I got a call from my doctor.

What is it? A tumor?

However, in the presence of the tumor, now there are molecular scissors that can break this molecule apart right there at the cleavable site.

So on. So this example...

In other machine learning problems we will often have more features, and my friends that work on this problem, they actually use other features like these, which is clump thickness, the clump thickness of the breast tumor.

But even before I peel that muscle away, you saw that there was a tumor underneath.

Previous lipo and breast augmentation with complications of seroma and scarring.

We r emoved a tumor fr om her brain.

Tom had a brain tumor.

When we put these two probes together ...

As you can see, it's blocking her vision.

The tumor's still small.

But there's been a significant reduction in the tumors.

And now, boom, the tumor labels itself and it gets fluorescent.

My breasts are tender.

The molecule passes through and gets excreted.

That's impressive.

How about more targeted therapies for cancer?

Here is an example.

First of all, these electrodes are noninvasive.

You have a secondary tumor in the long bone of your arm

Not unless a tumor is made of glass.

But this technology could not find tumors when they're small, and finding a small tumor is critical for survival.

This is a tumor dark, gray, ominous mass growing inside a brain.

I'm always interested to look at DNA and mutations.

I extracted DNA from them.

I want to look at the blood vessels. I want to look at the liver.

Henriapi illustrates this phenomenon as well.

Parasagittal astrocytoma.

He has extensive experience in tumor removal.

Only instead of, you know, good and bad cancer tumors or benign or malignant tumors we instead have diabetes or not and so we will use that as a supervised, we will solve that as a Supervised Learning problem just like we did for the breast tumor data.

So, if angiogenesis is a tipping point between a harmless cancer and a harmful one, then one major part of the angiogenesis revolution is a new approach to treating cancer by cutting off the blood supply.

But what we can see is, by month six, the tumor has responded and it's begun to melt away.

So here's a tumor.