Detecting Cancer From a Drop of Blood (The Anti-Theranos)

Summary

This video discusses groundbreaking advancements in cancer detection using AI and a single drop of dried blood. Scientists in China have developed a test that could revolutionize early cancer diagnosis, overcoming previous failures like Elizabeth Holmes' Theranos. The script explores the complexity of cancer, its increasing rates, and the potential of metabolites as stable biomarkers. It highlights how AI can detect patterns invisible to humans, offering hope for a cheap, accessible diagnostic tool that could save millions of lives.

Takeaways

• 🧬 Scientists in China have developed a test using artificial intelligence that can detect cancer from a single drop of dried blood.
• 🚫 The script references the Theranos scandal, highlighting the challenge of turning innovative ideas into practical solutions without compromising ethics and accuracy.
• 📈 The script discusses the rising global cancer rates, predicting a 77% increase by 2050, emphasizing the urgent need for better diagnostic tools.
• 🔍 The video mentions the potential of metabolites as a new type of biomarker for cancer detection, a shift from traditional markers like microRNAs and proteins.
• 🤖 AI's role in cancer detection is highlighted, showing its ability to recognize patterns that humans cannot, thus improving the accuracy of cancer diagnosis.
• 🌐 The script points out the accessibility and affordability of the new blood test, suggesting it could be a game-changer for global cancer detection, especially in underserved areas.
• 🔥 The video describes the process of using mass spectrometry and machine learning to analyze metabolic changes from dried blood spots for cancer detection.
• 💉 The script mentions that the new test requires only 0.05 mL of blood, making it less invasive and more cost-effective compared to existing diagnostic methods.
• 🌡 The stability of metabolites during transportation and temperature changes is noted, which is a significant advantage for the test's reliability and practicality.
• 🌐 The script touches on the global disparity in cancer detection, with only 5% of cancer spending reaching countries where 80% of cases occur, underlining the potential impact of this new test.
• 🔮 The video concludes with optimism about the future of AI in medicine, suggesting that early and affordable cancer detection could soon be a reality.

Q & A

What is the potential breakthrough in cancer diagnosis mentioned in the video?

The video discusses a breakthrough in cancer diagnosis using a single drop of blood and artificial intelligence developed by scientists in China.

Who is Elizabeth Holmes and what was her original idea related to blood testing?

Elizabeth Holmes is the founder of Theranos, a company that claimed to be able to diagnose a range of diseases from just a pinprick of blood. However, it was later revealed to be a high-profile fraud case.

Why is cancer considered a complex disease?

Cancer is complex because it comprises over 200 distinct diseases, each with its subtypes that vary in behavior and genetics.

What is the estimated increase in new cancer cases by 2050 compared to 2022?

By 2050, there will be an estimated 35 million new cancer cases, which is a 77% increase from the 20 million estimated cases in 2022.

What is a biomarker and why are they important in cancer detection?

Biomarkers are biological molecules within the body that indicate whether everything is functioning well or not. They are important in cancer detection because they can provide insights into the presence and progression of cancer.

What is a metabolite and how does it relate to cancer detection?

Metabolites are byproducts of metabolism that can offer insights into the physiological or pathological state of the body. They are considered a promising biomarker for cancer detection because they can indicate cancer progression.

How does the new AI blood testing method work for detecting cancer?

The new method uses mass spectrometry along with machine learning to analyze metabolic changes from dried blood spots, allowing for the detection of cancer with high accuracy.

What are the advantages of using dried blood spots in cancer detection tests?

Dried blood spots are advantageous because they remain stable across a range of temperatures, are less expensive, faster to perform, take up less physical space, and are easier to transport, making the test accessible to underserved populations and less developed countries.

What is the potential impact of this AI blood testing method on global cancer detection?

The AI blood testing method could make cancer detection more accessible, affordable, and accurate globally, especially in underserved areas and less developed countries where cancer is often missed.

What other AI-based cancer detection methods have been mentioned in the video?

The video also mentions the use of AI to analyze cell-free DNA (CF DNA) fragments released by dying cancer cells to detect liver cancer, and a UK study that theoretically makes it possible to detect cancer 7 years before traditional diagnosis.

How does the video address the issue of media bias in reporting on scientific breakthroughs?

The video introduces Ground News, a website that provides a data-driven, objective way to read the news, allowing readers to compare different articles, understand media bias, and get a full picture of various perspectives on scientific breakthroughs.

Outlines

00:00

🧬 AI and Blood Testing for Cancer Diagnosis

The script introduces a revolutionary approach to cancer diagnosis using artificial intelligence (AI) and blood testing. It references the infamous case of Elizabeth Holmes and her company Theranos, which promised blood tests with a single prick but turned out to be fraudulent. The video pivots to highlight how advancements in AI might turn such an idea into reality, with Chinese scientists developing a test that uses AI to detect cancer from a single drop of dried blood. The episode aims to explore how this technology could change cancer diagnosis and treatment, noting the complexity and increasing rates of cancer worldwide.

05:01

🔬 Metabolites: The New Biomarkers for Cancer Detection

This paragraph delves into the scientific aspect of metabolites, which are portrayed as the new frontier in cancer detection. It explains how metabolites, the byproducts of cellular processes, can serve as indicators of physiological or pathological states within the body. The script likens metabolites to exhaust smoke from a car, with AI acting as a sensor to analyze these biomarkers for signs of cancer. The study mentioned utilizes mass spectrometry and machine learning to scrutinize changes in metabolites from dried blood spots, achieving high accuracy in detecting cancer. The AI's ability to recognize patterns invisible to humans is emphasized, along with the advantages of using dried blood for its stability and cost-effectiveness, making it a promising tool for global health equity.

10:02

🌐 Ground News: Unbiased News Consumption

The final paragraph shifts focus to the importance of unbiased news consumption, introducing Ground News, a platform developed by a former NASA engineer. It offers a data-driven, objective way to read the news, using the example of how different news outlets cover the AI and cancer story. The platform provides tools to assess media bias and compare various articles on the same topic, helping users become critical thinkers. The script promotes a 40% discount on the Ground News Advantage subscription, which includes features like 'my news bias' and a 'blind spot feed' to ensure well-rounded news consumption.

Keywords

💡Artificial Intelligence (AI)

Artificial Intelligence refers to the simulation of human intelligence in machines that are programmed to think like humans and mimic their actions. In the context of the video, AI is pivotal as it is used to analyze and interpret complex metabolic data from blood samples to detect cancer with high accuracy. The script mentions how AI can recognize patterns that humans cannot, which is crucial in the early detection of diseases like cancer.

💡Cancer

Cancer is a group of diseases characterized by the uncontrolled growth and spread of abnormal cells. The video discusses the relentless nature of cancer, its complexity with over 200 distinct diseases, and the increasing rates of cancer worldwide. It emphasizes the importance of early detection and how advancements in AI and blood testing could revolutionize the fight against cancer.

💡Metabolites

Metabolites are the intermediates and products of metabolism. They are the substances that our body produces during the chemical processes that occur within us. In the video, metabolites are highlighted as a new focus in cancer research, offering insights into the physiological state of the body and being a potential biomarker for cancer detection through AI analysis.

💡Biomarkers

Biomarkers are biological molecules found in blood, other body fluids, or tissues that are a sign of a normal or abnormal process, or of a condition or disease. The script discusses the search for reliable biomarkers in cancer detection and how metabolites have emerged as promising candidates for this purpose.

💡Dried Blood Spots

Dried Blood Spots (DBS) are samples of blood that have been spotted onto a paper card and dried. The video explains that DBS can be analyzed immediately, remain stable across a range of temperatures, and are cheaper to transport, making them ideal for the AI-based cancer detection method discussed.

💡Mass Spectrometry

Mass spectrometry is an analytical technique used to identify the chemical composition of a substance by measuring the mass-to-charge ratio of its ions. In the context of the video, mass spectrometry is used in conjunction with machine learning to analyze metabolic changes from dried blood spots for cancer detection.

💡Early Detection

Early detection refers to the identification of a disease or condition at an early stage, often before symptoms appear. The video emphasizes the importance of early cancer detection for increasing survival rates and how AI and blood testing advancements could make this possible.

💡Elizabeth Holmes

Elizabeth Holmes is the founder of the now-defunct health technology corporation Theranos, which claimed to have developed a method for blood tests requiring only a few drops of blood. The video references her as an example of the potential and pitfalls of health technology innovation, contrasting her story with the current advancements in AI and blood testing for cancer detection.

💡Underserved Populations

Underserved populations refer to groups that lack access to healthcare services and resources. The script mentions that the new AI-based blood test for cancer detection could be particularly beneficial for these populations, offering a less expensive and more accessible diagnostic tool.

💡Clinical Trials

Clinical trials are research studies that involve human volunteers to test new medical interventions. The video acknowledges that despite the promising advancements, there is still a need for major clinical trials to validate the AI-based cancer detection methods before they can be widely implemented.

💡Cell-Free DNA (cfDNA)

Cell-Free DNA refers to DNA that is found in blood or other body fluids and is not contained within a cell. The script mentions a study where AI was used to analyze cfDNA fragments to detect liver cancer, illustrating another application of AI in cancer detection.

Highlights

Scientists in China have created a test using artificial intelligence that can detect cancer using a single spot of dried blood.

The idea of diagnosing diseases from a pinprick of blood was proposed by Elizabeth Holmes with Theos but was deemed impossible at the time.

Advancements in AI may now make early and accurate cancer detection from a drop of blood possible.

Cancer is complex, with over 200 distinct diseases, each with varying subtypes in behavior and genetics.

By 2050, there will be an estimated 35 million new cancer cases, a 77% increase from the 20 million estimated cases in 2022.

Early onset cancers have increased by 79% for those under 50, indicating a shift in the demographics of the disease.

80% of Asian-American women with lung cancer have never smoked, and they are 1.5 to 2 times more likely to be diagnosed than any other racial group.

Currently, there are no reliable or standalone blood tests that can diagnose deadly diseases such as pancreatic, gastric, and choral cancer.

A new study suggests a method to detect the world's deadliest cancers with an accuracy of 82% to 100% of the time.

Metabolites, the focus of modern cancer research, offer insights into cancer progression and are considered the best biomarker for cancer.

AI can recognize patterns that humans can't see and extract useful information from them, improving diagnostic accuracy.

The new non-invasive test requires just 0.05 mL of blood and is less expensive, faster, and more accessible than current tests.

Dried blood spots are stable across a range of temperatures and are cheaper to transport, making the test accessible to underserved populations.

Metabolites stay stable during transportation and temperature changes, providing highly accurate diagnostics.

AI has been used to detect more than 80% of liver cancers by analyzing cell-free DNA fragments released by dying cancer cells.

UK scientists have theoretically made it possible to detect cancer 7 years before traditional diagnosis using AI.

Ground News provides a data-driven, objective way to read the news, helping readers understand different perspectives and media bias.

AI's application in medical diagnostics shows promise, with the potential to save millions of lives through early detection of cancer.

Despite the hype and skepticism around AI, its use in pattern recognition and prediction in data can be highly beneficial in many fields.

Transcripts

this video is brought to you by ground

news Dr Malik Marshall is here with us

now doctor we could soon diagnose cancer

with a single drop of blood now can you

believe that these are scientist in

China that created a test using

artificial intelligence that can detect

cancer using one spot of dried blood hi

welcome to another episode of Cold

Fusion we all remember Elizabeth Holmes

and Theos it was one of the most

high-profile fraud cases in recent

memory her original idea was simple from

just a pin PRI of blood patients could

get a diagnosis of a range of diseases

and when you think about it it was a

pretty good idea the only problem was it

was imposs Holmes with her delusions of

grandeur thought that she could be the

next Steve Jobs and change the world the

end result was investors that were

scammed out of tens of billions and the

health of innocent patients put at risk

but what if she was just a little ahead

of her time what if thanks to the

advancements of AI The Impossible Is Now

possible multiple Studies have shown

very promising signs the detection of

cancer all from a drop of blood in this

episode we'll see how advancements and

blood testing as well as AI may change

the fight against cancer

forever you are watching T Fusion

[Music]

TV cancer is a terrible thing it rips

families apart and ravages the bodies of

those suffering it's been causing

destruction since its earliest

documentation in ancient Egypt yes

fossilized bone tumors have been found

from 5,000 years ago and you'd think

after all of this time we'd have figured

out something by now but the hard truth

is this cancer is so relentless because

it's complex comprising of over 200

distinct diseases each with its subtypes

that vary in behavior and genetics for

this reason despite technological

advancements Innovation and Relentless

research cancer rates are still on the

rise by 2050 there will be an estimated

35 million new cancer cases and that's a

77% increase from the 20 million

estimated cases in 2022 in a few short

years for the first time in history

cancer is on track to be Humanity's

number one killer but are we just seeing

more cancer cases because our diagnosis

tools have gotten better it's possible

but recent Trends are screaming

something else cancer has always been

thought of as a disease of the old but

what if I told you that that's changing

early onset cancers have increased by

79% for those under 50 so something else

is going on is it obesity

ultra-processed diets stress even toxic

forever chemicals like pasas have been

blamed and then there's the very strange

phenomenon of specifically Asian

non-smoking women ending up with lung

cancer at 150 to 200% of the rate of

other communities this afternoon we want

to shed light on a condition impacting

more and more Asian-American women and

it remains a mystery as to why and now a

Bay Area based study finds that 80% 80%

of Asian-American women with lung cancer

never smoked they're also one and a half

to two times more likely to be diagnosed

than any other racial group scientists

really don't have an answer for any of

it but there's hope so let's take a look

at

that where you live can determine your

odds of survival the best thing to do

for cancer is to detected early as that

dramatically increases our chances of

survival but there's a problem there are

currently no reliable or Standalone

blood tests that can diagnose deadly

diseases such as pancreatic gastric and

choral cancer and this is why the lack

of approved biomarkers biomarkers are

biological molecules within the body

that tells doctors if everything's

functioning well or not instead doctors

have to rely on Imaging or surgical

procedures each Limited in their own way

but what if there was a way to detect it

extremely early even before symptoms

what if it could be done in such a way

that it's cheap and affordable and

within the reach of worldwide

communities that sounds insane but we're

actually on the verge of it happening

today in the past few months and years

there's been a silent Revolution going

on with AI blood testing and cancer this

new research paper says that there could

be in fact a way to detect pancreatic

gastric or chexal cancer and get the

diagnosis right 82 2% to 100% of the

time they've come up with a method to

detect the world's deadliest cancers

with staggering accuracy so how did they

achieve such remarkable

accuracy to fully answer that question

we first need to take a step back and

understand some background here and it

all starts with

metabolites modern cancer research

focuses on the discovery of biomarkers

but recently scientists have zeroed in

on a new one metabolites without getting

to into the Weeds on this here's the

tldr version otherwise this video could

be hours long think of it like this our

genetic makeup combined with the

proteins in our body give insights into

our normal physiological state or an

abnormal pathological state are things

in the body going well or poorly

metabolites can offer insights into

cancer progression and scientists have

found this measure to be probably the

best biomarker for cancer but how does

it

work imagine your body is a petrol car

that's running the exhaust smoke is the

metabolite and imagine that there's a

sensor that measures the composition of

the exhaust smoke the sensor is the AI

in our example if the engine isn't quite

right then there can be some telltale

signs such as changes in the composition

of the exhaust Smoke by using the sensor

in this crude example how do we

interpret the data and tell exactly

what's wrong with the car just by the

sample of exhaust smoke you'd have to

build a system that can associate slight

variations in Smoke composition with a

particular engine problem well this is

where the AI comes in in this example

the AI is the sensor which is reading

the exhaust smoke which are the

biomarkers and in the world of cancer

research recently using artificial

intelligence like this is becoming more

common but the thing is targeting

metabolites aren't new it dates back to

1947 and the work of Dr Sydney Farber

known as the father of modern

chemotherapy however really getting to

the bottom of cancer cell metabolism has

always been impeded by the technological

capabilities it was almost impossible to

actually handle and analyze the chemical

complexity of all those metabolites and

their patterns it's just way too hard to

do by hand or by most hard-coded

[Music]

programs with that background being

understood now we can get into the study

in the study scientists used a special

kind of mass spectrometry they used this

along with machine learning to analyze

metabolic changes from dried blood spots

the system was able to analyze complex

metabolic data with

accuracy in previous studies artificial

intelligence was found to be almost

twice as accurate in assessing the

aggressiveness of certain kinds of

cancer so this is AI in action what do

we see here AI circled this area in red

and said this area has a dangerous

pattern it looks like cancer might come

up 2 years later there's a cancer right

in that same spot where the AI predicted

it would be and right there is the Crux

of it AI can recognize patterns that

humans can't see and extract useful

information from it it's pretty ironic

when you look at all the AI scams going

on there are good sides too and plus in

this case the more samples the AI

processes the more accurate the

predictions and there's more good news

this method is less cumbersome and also

cheap this new non-invasive test would

theoretically require just 0.05 mL of

blood now if this sounds familiar it's

because we've heard similar promises

before so what's different this time

dried blood for starters dried blood can

be analyzed right away they remain

stable across a range of temperatures

and are cheaper to transport they found

the blood spot test had greater accuracy

than current tests youing using whole

blood in fact the spot test was more

heat stable it was less expensive faster

to perform took up less physical space

and was easier to transport making the

test particularly accessible to

underserved populations and less

developed countries where cancer are

often missed but it sounds remarkable

and so much less expensive than all

these other diagnostic tests that people

are going through right now incredible

Okay the reason you're only hearing

about this now is because previously

common cancer biomarkers such as

micrornas and proteins couldn't handle

the drying process and this is because

it messed up the test accuracy but when

scientists shifted their focus to

metabolic biomarkers they discovered

that metabolites stayed stable during

transportation and temperature changes

and they still provided highly accurate

Diagnostics we're still missing the mark

on a diagnostic Solution that's

accessible to all only 5% of global

cancer spending reaches the countries

where 80% of the cancer cases are and

less than 30% of poorer income countries

have the diagnostic tools they need most

cancer detection methods are costly and

require specialized staff but even in

countries with access to such methods

there are significant drawbacks pet

scans can Overlook small tumors and also

interpretations can vary from Doctor to

doctor biopsies are of course invasive

and hard to come by globally regular

blood tests lead to many false positives

and endoscopies come with their own side

effects so with this new Innovative

Fusion of dried blood serum metabolite

biomarkers and artificial intelligence

you can see why heads are turning and

ears are perking up and this will be

especially beneficial for underserved

areas with limited resources incredibly

this is not the first time AI has been

used in this exact

way back in 2022 researchers at John

Hopkins kiml Cancer Center found that AI

could detect more than 80% of liver

cancers however in this study of 724

people a different biomarker was used

DNA fragments you see when cancer cells

die they release DNA fragments into the

bloodstream this is known as cell-free

DNA or CF DNA CF DNA can come from both

healthy and cancerous cells but

cancerous cells exhibit irregular

patterns of fragmentation so the

researchers used machine learning to

examine millions of of these fragments

and then amazingly they could identify

the subtle patterns associated with

cancer awesome stuff another recent

breakthrough has been achieved by UK

scientists their work theoretically

makes it possible to detect cancer 7

years before traditional diagnosis but

it's also interesting to see how the

news is covered and that's where today's

sponsor ground news comes in ground news

is a website developed by a former NASA

engineer she was on a mission to give

readers an easy datadriven objective way

to read the news using this cancer story

as an example we can see that 26 news

outlets covered around the world their

biased distribution chart shows me the

political leanings of those news outlets

and I can get a summary of how the issue

was being framed the left tended to use

heavier language like warning signs and

deadly while the riot didn't have such

language in this case for more

politically charged stories ground news

is extremely helpful to get a full

picture of the different perspectives

scrolling down I can compare every

single article on the topic with

convenient tags showing me cont text

about the source like how factual it is

and who owns it ground news is a

fantastic tool for getting International

views sifting through misinformation and

identifying media bias they provide all

the tools you need to be a critical

thinker today we're offering 40% off

Advantage subscription this comes with a

new feature called my news bias so you

can even check how your news viewing

habits change over time one of my

favorite features is the blind spot feed

this shows you stories that are

underreported by one side of the

political Spectrum for example if you

lean left you probably miss the story

about the largest number of people

crossing the English Channel in a day

this year go to grounds.com coldfusion

to get started thanks okay now back to

the

video so while this is all very

encouraging we have to take a look at

the whole picture we're still

realistically a few years away from

these AI powered techniques being put to

work to up those diagnosis rates major

clinical trials require a lot of money a

lot of trial patients and a lot of

regulatory Cycles I still think it's

clear that there's New Hope in the fight

against cancer it's a multi-pronged

approach that combines new cancer

Discovery utility practicality and the

most powerful new technology thinking

about it with AI bolted onto the side of

human Ingenuity we could see a new error

in the medical field now everyone could

have access to a cheap and easy method

to detect cancer just walk into your

local doctor's office take a pin prick

of blood and get your results the

algorithms will only get better and

ultimately millions of lives could be

saved with early detection how cool

would that be so at this point I think

it's fairly normal to have mixed

feelings about AI on the one hand it's

overplayed overhyped there's fraud lying

and employment theft in some sectors but

on the other hand it can do some amazing

things basically anything that involves

patent recognition and prediction

through data can be enhanced through Ai

and that's a lot of stuff you just have

to think outside the box sometimes times

all in all I don't think anyone can be

mad at this application it's very

interesting isn't it anyway thanks for

watching my name is toogo and you've

been watching cold fusion and I'll catch

you again soon for the next episode

cheers guys have a good

[Music]

one cold fusion it's me thinking

[Music]

you something to me something