(Laughter)

Ever since then, a bunch of people have come up to me and asked,

"How on Earth could a 15 year old have come up with a new way

to detect pancreatic cancer?"

My answer, "A ton of hard work, a year and a half to be precise,

and a ton of failures."

Recently, I developed a novel paper sensor for the detection

of pancreatic, ovarian, and lung cancer.

It is 168 times faster, over 26,000 times less expensive

and over 400 times more sensitive than the current gold standard of detection.

The best part is --

(Applause)

the best part, it costs 3 cents and takes 5 minutes to run.

(Applause)

It all began one day when I decided to go online and start researching

statistics on pancreatic cancer.

We had recently lost a close family friend who was like an uncle to me

who had succumbed to the disease of pancreatic cancer.

What I found was eye-opening.

Over 85% of all pancreatic cancers are diagnosed late

where a patient has less than 2 percent chance of survival.

That means less than 2 people out of every 100 [survive].

In addition, it has an abysmal 5 year survival rate.

Only 5.5 percent of people will survive after 5 years.

The average life span of someone with pancreatic cancer is 3 months.

One of my dad's friends actually suffered of pancreatic cancer,

and a week later he was dead.

So I was wondering, why we are so bad at detecting pancreatic cancer.

What I found was eye-opening and shocking to me.

Our "modern medicine" is a 60 year old technique.

It is highly outdated and grossly inaccurate.

It misses over 30 percent of all pancreatic cancers.

In addition, it's pricey.

It costs 800 dollars per test and it's not covered by insurance.

So, it's not an option for low income patients.

In addition, pancreatic cancer is a non symptomatic disease.

That means that all of its symptoms are really general

such as abdominal pain, jaundice.

So, a doctor can't easily diagnose it.

Then I started making a scientific criteria,

that I would imagine a sensor that was optimal would have.

It would have to be simple, sensitive, selective, rapid, inexpensive,

and minimally invasive to a patient.

I was pretty confident that I could create such a sensor, but I wasn't quite sure how.

Then, I started doing a bit more research

and I found out why such a technological advancement hadn't been made.

What I found is that, due to the daunting nature of discovery,

no work has really been done on this.

What is happening with pancreatic cancer when you diagnose it,

you are looking for a cancer biomarker

or a protein that's found at higher levels in your blood stream.

This sounds really straightforward, but it is anything but.

You see, you have all this healthy blood, liters and liters of healthy blood.

But, you are looking for this tiny increase in this tiny amount of protein.

That's next to impossible.

Essentially, what you are doing is you are looking for a needle in a haystack.

But worse, you are looking for a needle in a stack of [nearly identical] needles.

So then, what I did, is I began researching because I had to find some target to look at.

I started actually with a database of over 8,000 different proteins

found in pancreatic cancer.

Luckily, on the 4,000th try, I finally hit gold.

I found this protein I could use.

Its name was mesothelin.

It is just your regular protein

unless you have pancreatic, ovarian, or lung cancer.

In which case, it's found at highly expressed level,

at highly over expressed like really high levels in your blood stream.

Then, the key about this protein is that it's found early in the disease

when a patient has close to 100% chance of survival.

So, if I could detect this protein,

then I could hopefully cure pancreatic cancer, basically.

Then, I shifted my focus to trying to detect the protein because that was the big question.

My breakthrough came in the most unlikely of places.

It came in high school biology class -- the absolute abhor of innovation.

(Laughter)

I basically smuggled in this article on single walled carbon nanotubes

I had been dying to read.

A single walled carbon nanotube is essentially an atom-thick tube of carbon.

That's -- just imagine a really long pipe.

It is one 150th of the diameter of your hair.

And it has these amazing properties.

They are super, super cool.

They are like the superheros of material science.

Then, I was trying to roll over this concept of -- we were learning about -- antibodies.

Antibody is basically a lock and key molecule

that attaches specifically to a certain protein, in this case, the mesothelin.

I was trying to combine that specific reactivity

to how carbon nanotubes are really sensitive to their network

of the 3 dimensional structures of their network.

Then, it hit me.

What I could do is I could put an antibody in this network

such that would react specifically to the mesothelin.

Then, also I would change its electrical properties based on the amount of mesothelin,

enough so that I could measure it with the 50 dollar Home Depot ohmmeter.

So, pretty easy.

Just as I had this epiphany, my biology teacher storms up to me,

because she spots me reading this article, snatches it out of my hand,

because I was supposed to be writing an essay,

then, storms off and gives me a lecture.

After class, I finally convinced her after a huge lecture

on how I should respect her in her class --

I finally got my article back because that is all I really wanted from her.

(Laughter)

Then, what I did is I began researching this promising idea.

Then, I needed a lab space

because you can't do cancer research on your kitchen countertop.

(Laughter)

Basically, what I did is I wrote up a budget, a timeline, a procedure, and a materials list

so all the professors I emailed knew that I meant business.

So, then, what happened is I emailed 200 different professors

at the National Institutes of Health and Johns Hopkins University.

Basically, anyone who had anything to do with pancreatic cancer.

Then, I was kind of expecting to sit back and wait for positive emails to flow in

and I would get a pick and choose.

(Laughter)

Then, reality took place.

Over the course of a month, I got 199 email rejections.

One of them went as far as to systematically pop a hole in each part of my procedure.

So, it was a bit depressing.

But, there was 1 lukewarm maybe professor.

I finally tracked him down, after 3 months, nailed down an interview.

I go in with my knowledge of 500 plus journal articles I have read.

We start the interrogation.

Because what happens is over the course of this hour long interview

he calls in more and more experts, trying to pop holes in my solution.

I sit through all of it and I answer all of his questions.

I guessed on a few of them. (Laughter)

But, the interrogation paid off.

I got the lab space I needed.

Then, I started on a 7 month long journey in order to finally find the solution.

It seemed at first nothing was working.

Everything was really screwed up and there were millions of holes in my procedure.

Over the course of 7 months,

I slowly, painstakingly filled each and every one of those.

At the end, I ended up with the paper sensor

that could detect a 100% of all pancreatic, ovarian, and lung cancers.

(Applause)

But, I've learned a really important lesson over the course of my journey.

What I've learned is that through the Internet, anything is possible.

Theories can be shared and you don't have to be a professor

with multiple degrees to have your ideas valued.

Regardless of your gender, your age your ethnicity, regardless of anything,

it's just your ideas that count.

To me, that's all that really matters.

"Redefining relevance" for me is looking for new ways to use the Internet.

We really don't want to see your duckface pictures.

(Laughter)

Instead, you could be changing the world with the Internet.

You could help detect pancreatic cancer.

So, if I could detect pancreatic cancer,

just imagine what you could do.

Thank you.

(Applause)