Future Medicine: Modern Informatics | Richard Frackowiack | TEDxYouth@Zurich

Summary

The speaker, a retired brain doctor, emphasizes the need for a revolution in brain medicine, criticizing the stagnant approach that has persisted for over a century. He discusses the establishment of the Human Brain Project, a massive, billion-euro initiative spanning 10 years and involving 70 principal investigators across 23 countries. The project aims to leverage informatics to understand brain organization and disease mechanisms, seeking better treatments. The speaker highlights the underutilization of medical data and the importance of data integration and federation to create 'disease signatures' for more accurate diagnoses. He also addresses ethical concerns, advocating for a distributed approach to data analysis that respects privacy and empowers patients to contribute their data for medical advancement.

Takeaways

• 👨‍⚕️ The speaker is a retired brain doctor who is concerned about the lack of progress in brain medicine and the current methods that have remained largely unchanged for over a century.
• 🧠 There is a significant gap in understanding the brain and its diseases, with only a 30% accuracy rate in diagnosing dementia, even in renowned medical centers.
• 💡 The speaker emphasizes the need for a new approach, integrating computer science and informatics into brain medicine to improve diagnostics and treatment.
• 🌟 The Human Brain Project was initiated with a billion Euros investment, aiming to understand brain organization and disease mechanisms over a span of 10 years.
• 🏥 The project recognized the vast amount of clinical data available in hospitals, which is currently underutilized and could be a valuable resource for research.
• 🔒 Privacy and security are paramount, with strict measures in place to protect patient data, ensuring it is used ethically and responsibly.
• 🧬 The integration of clinical and research data is crucial for advancing medical knowledge, despite the challenges of data accessibility and sharing.
• 📊 The use of advanced imaging and data analysis can reveal disease signatures, allowing for more accurate diagnoses, such as identifying Alzheimer's disease with 95% accuracy.
• 🌐 The project aims to create a medical informatics platform to federate data across multiple hospitals, enhancing the understanding and treatment of brain diseases.
• 🔑 The speaker calls for a cultural shift in medicine, where doctors embrace informatics and patients advocate for the use of their data to improve medical research and treatment.

Q & A

What is the main concern expressed by the retired brain doctor in the script?

The main concern is the outdated and ineffective methods in treating brain-related diseases, such as psychiatric disorders and dementia, which have seen little improvement over the past 150 years.

Why does the speaker believe that the integration of computers and informatics is crucial for the advancement of brain medicine?

The speaker believes that the integration of computers and informatics is crucial because it can help analyze and utilize the vast amounts of clinical and research data that are currently underutilized, leading to better understanding, diagnosis, and treatment of brain diseases.

What is the Human Brain Project mentioned in the script, and what are its goals?

The Human Brain Project is a large-scale scientific research initiative funded with a billion Euros over 10 years. Its goals include understanding how the brain is organized, the mechanisms underlying diseases, and finding better treatments by integrating data from various sources.

How does the speaker describe the current state of clinical data usage in hospitals?

The speaker describes the current state of clinical data usage as inefficient, with only 5 to 8% of data being used for patient follow-ups, while the rest remains unused, except by lawyers for malpractice cases.

What is the significance of 'disease signatures' in the context of the script?

Disease signatures refer to the patterns identified through data analysis that can help in diagnosing diseases with high accuracy. In the context of the script, they are used to diagnose Alzheimer's disease with 95% accuracy, compared to the traditional 70% accuracy of hospital clinical work-ups.

What is the speaker's vision for the future of medical diagnosis with the help of informatics?

The speaker envisions a future where medical diagnosis is not solely based on patient interviews and examinations but also incorporates a wide range of data to create disease signatures, allowing for more accurate, data-driven diagnoses.

How does the speaker propose to address the ethical issues of privacy in medical data usage?

The speaker proposes a strategy that avoids aggregating data in large warehouses. Instead, they suggest developing software to distribute questions to hospitals, which would then provide only the necessary information without revealing individual data, thus preserving privacy.

What is the role of the patient in the envisioned medical informatics revolution according to the speaker?

The speaker sees the patient as a key driver of the medical informatics revolution, encouraging them to consent to the use of their data for medical research and to demand that their doctors embrace the informatics revolution.

What are the challenges faced by the Human Brain Project in terms of data usage?

The challenges include the incomplete, unstructured, and non-standardized nature of clinical data, as well as the reluctance of pharmaceutical companies and researchers to share their data due to commercial interests or personal recognition.

How does the speaker plan to demonstrate the effectiveness of the medical informatics platform?

The speaker plans to demonstrate the platform's effectiveness by implementing it in a network of six hospitals across different cities and then, if successful, tendering it to European companies for wider implementation in hospitals.

Outlines

00:00

🧠 The Need for a Paradigm Shift in Brain Medicine

The speaker, a retired brain doctor, expresses disappointment in the current state of brain medicine, which has seen little change in 150 years. He emphasizes the need for a new approach, particularly with the integration of computer science and informatics. The speaker highlights the lack of progress in treating psychiatric diseases and dementia, noting that even in prestigious medical centers, the cause of dementia is misunderstood 30% of the time. He introduces the Human Brain Project, a massive, billion-euro initiative spanning 10 years and involving 70 principal investigators across 23 countries, aimed at understanding brain organization and disease mechanisms to develop better treatments.

05:01

🌐 Leveraging Data and Informatics in Brain Medicine

The speaker discusses the vast amounts of clinical and research data that are currently underutilized in hospitals and pharmaceutical companies. He argues for the need to integrate and federate this data to unlock its potential for medical advancement. The speaker provides examples of how data integration can lead to disease signatures, which are patterns that can significantly improve diagnostic accuracy, such as increasing Alzheimer's diagnosis accuracy to 95%. He also stresses the importance of using these disease signatures to transform diagnostics, moving from patient interviews and exams to a more data-driven approach that incorporates a wide range of information.

10:03

🔬 Advancing Medical Understanding Through Disease Signatures

The speaker explains how disease signatures can help in understanding the progression of brain diseases like dementia and how they can provide real prognosis for patients. He discusses the importance of using computing to analyze data and identify these signatures, which can reveal different mechanisms at play in various types of dementia. The speaker also touches on the ethical considerations of data privacy and the strategy to avoid centralizing data in large warehouses, instead proposing a distributed approach where questions are sent to hospitals to retrieve necessary information without compromising individual privacy.

15:04

🏥 Implementing Medical Informatics for Future Healthcare

In the final paragraph, the speaker outlines plans to develop a medical informatics platform and create a network of hospitals to demonstrate its effectiveness. He discusses the importance of patient consent and the ethical use of data for medical research, emphasizing the need for a balance between public health and privacy rights. The speaker concludes by urging the audience to embrace the informatics revolution in medicine, advocating for a future where medical practice is enhanced by data-driven insights and technologies.

Keywords

💡Brain doctor

A brain doctor, also known as a neurologist, is a medical professional who specializes in diagnosing, treating, and managing disorders of the nervous system, including the brain. In the context of the video, the speaker identifies himself as a retired neurologist, emphasizing his expertise and experience in the field of brain medicine. His disappointment with the current state of neurological treatments and his optimism for change reflect his professional background and personal commitment to improving brain healthcare.

💡Psychiatric disease

Psychiatric disease refers to a range of mental health disorders that affect an individual's mood, thinking, and behavior. The video discusses the long-term suffering and lack of effective cures for young people with psychiatric diseases, highlighting the need for a more advanced approach to understanding and treating these conditions. The speaker's focus on this issue underscores the importance of addressing the mental health crisis and the potential role of informatics in finding solutions.

💡Dementia

Dementia is a term used to describe a group of symptoms associated with a decline in cognitive abilities severe enough to interfere with daily life. The video mentions the challenges faced by individuals and their families as they cope with the progressive nature of dementia, and the speaker's concern about the current misdiagnosis rate. The integration of data and informatics in the video is presented as a potential solution to improve the accuracy of dementia diagnosis and treatment.

💡Informatics

Informatics, in the context of the video, refers to the integration of computer science and information technology to manage, analyze, and interpret complex data sets, particularly in the medical field. The speaker argues that the application of informatics to medicine, specifically neurology, is crucial for advancing the understanding of brain diseases and developing more effective treatments. The video emphasizes the potential of informatics to revolutionize traditional medical practices.

💡Human Brain Project

The Human Brain Project is a large-scale research initiative aimed at understanding the brain and its diseases by simulating the brain's functions using supercomputers and analyzing vast amounts of data. The video describes the project as a collaborative effort involving multiple investigators, institutions, and countries, with the goal of uncovering the mechanisms of brain diseases and improving diagnostic and treatment methods. The project exemplifies the application of informatics and data analysis in neuroscience.

💡Data integration

Data integration is the process of combining data from different sources and formats into a unified view. In the video, the speaker discusses the importance of integrating clinical and research data to create a comprehensive understanding of brain diseases. By integrating data from various sources, such as patient records, genetic information, and imaging scans, researchers can identify patterns and 'disease signatures' that can lead to more accurate diagnoses and personalized treatments.

💡Disease signatures

Disease signatures refer to the specific patterns or combinations of symptoms, genetic markers, and biological indicators that are characteristic of a particular disease. The video highlights the discovery of disease signatures for Alzheimer's disease and other forms of dementia through data integration and analysis. These signatures can help in diagnosing diseases with higher accuracy and understanding their progression, which is crucial for developing targeted treatments.

💡Neuroimaging

Neuroimaging is a branch of medical imaging that focuses on the structure and function of the nervous system, particularly the brain. The video mentions the use of neuroimaging data from the Alzheimer's Disease Neuroimaging Initiative, which has made its data publicly available for research purposes. The integration of neuroimaging data with other types of medical information is presented as a powerful tool for identifying disease signatures and advancing the understanding of brain diseases.

💡Ethical issues

Ethical issues in the video pertain to the responsible handling of patient data, ensuring privacy, and obtaining informed consent for research. The speaker emphasizes the importance of addressing these concerns when using informatics in medicine. The proposed solution involves developing software that allows for distributed data analysis without centralizing patient information, thus preserving privacy and adhering to ethical standards.

💡Medical informatics platform

The medical informatics platform mentioned in the video is a software system designed to facilitate the secure and efficient analysis of medical data across multiple hospitals. The platform aims to enable data integration and analysis without compromising patient privacy, supporting the development of disease signatures and improving diagnostic accuracy. The speaker's vision is to demonstrate the platform's effectiveness in a network of hospitals and then roll it out more broadly to enhance medical care.

Highlights

The speaker is a retired brain doctor and an optimist, despite being disappointed with the current state of brain medicine.

Brain medicine has remained largely unchanged for 150 years, with limited success in treating psychiatric diseases and dementia.

The speaker emphasizes the need for a new approach to medicine, particularly in the field of brain health.

The Human Brain Project was initiated with a billion Euros, aiming to understand brain organization and disease mechanisms over 10 years.

The project involves 70 principal investigators, multiple institutions, and participation from 23 countries.

A significant challenge is the underutilization of clinical data due to privacy concerns and lack of standardization.

The speaker calls for a change in the medical community's approach to data sharing and integration.

The potential of using advanced imaging techniques to understand brain function and disease is discussed.

The integration of clinical and research data can lead to the discovery of disease signatures, improving diagnostic accuracy.

The speaker highlights a 95% accuracy rate in diagnosing Alzheimer's disease using integrated data, compared to 70% in hospitals.

The concept of disease signatures is introduced as a way to visualize data and test hypotheses about diseases.

The importance of moving from molecular genetics to understanding higher cognitive functions is emphasized.

The speaker discusses the ethical issues of privacy and the strategy to avoid centralizing data in large warehouses.

A new software approach is proposed to distribute questions to hospitals and collect only necessary information.

The potential for patients to drive the informatics revolution in medicine is highlighted.

The speaker calls for a balance between the rights to public health and privacy when using medical data.

The development of a medical informatics platform is underway to demonstrate the effectiveness of the new approach.

The speaker concludes by urging the medical community and patients to embrace the informatics revolution for better healthcare.

Transcripts

[Music]

I'm a brain doctor or at least um I'm a

retired brain doctor and uh I'm a

disappointed person but as as you can

see from my socks I'm a I'm an

optimist and so as I was coming up to

retirement about 5 years before it in

fact I thought we got to change

things the medicine I'm talking about is

Medicine of the

brain

it's something that we do in the same

way we've always

done and we've been doing it for 150

years and we're still not very good at

it when it comes to the brain I mean you

all know people who are usually Young

Who suffer from psychiatric

disease usually for many years never

cured many of you now will know parents

grandparents who are losing their minds

dementing and then those last terrible

years where everyone has to look after

them or put them into a home or

whatever we're 30% wrong with the cause

of the dementia 30% even in you know

Mass General London Paris Lozan you know

the great centers of

Neurology so we've got to do something

about this and as we were thinking about

this a couple of mates and myself a

computer scientist a new scientist and

me a clinical scientist we thought that

the real thing that's

missing is the impact of computers and

informatics on

medicine in fact you the patients have

been much better at this than we the

doctors I mean you come in usually with

a pile of things like this off the

internet about your disease and then we

suddenly find ourselves being a bit

silly because we are not as up to date

as you

are so we set up this human brain

project a billion Euros 10 years 70

principal

investigators lots of Institutions 23

countries 10 years to try and understand

how the brain is

organized and to understand what disease

does to it what are the mechanisms

underlying disease how can we find

better

treatments so this is reasonably

big and the first thing we had to do was

find out what resources do we

have and then we thought to ourselves

which clicker are we going to use hey it

worked so the resources that we have are

data every hospital has Banks of data

about its patients for at least 10

years

secure uncorruptible

guarded against people looking into your

private

lives in fact unused you only use 5 to

8% of those data for followup patients

the rest is used by lawyers to attack

doctors who have done Mal practice and

think of the money that's used to build

data warehouses think of the energy

that's

used think of the lack of use and what

immense wealth of knowledge and real

wealth

has been just

LIF and it's your wealth because we in

Europe have socialized Health Systems

largely that means you the taxpayers are

paying for that and you're getting

nothing

back so on the medical side we said to

ourselves we need to solve this problem

we need to solve this problem with our

clinical data which there are vast

amounts in

Europe but clinical data which are not

complete not very structured not very

standardized a bit dirty with all this

protection around

them and also research data now this

comes from places like pharmaceutical

companies which hide them for commercial

reasons but if you go to the

pharmaceutical company and say how many

of your trials drug trials succeed and

they say oh we're lucky if they 5% 10%

succeed give us the data from the ones

that don't oh they hand it to

you you just need to think about it if

you go to researchers they say oh that's

my data I'm going to win my Nobel Prize

you say well hang on a moment it's not

the data that wins the Nobel Prize it's

your brain and your brain is clearly not

up to it if you're hiding your data

because you need to share it you need to

find more of it you need to get it all

together integrate it Federate it

integrate it try and do something with

it with the ideas you have not with

hiding data with passwords which you

forget and then corrupt and things like

that so so that's what we're going to

use and in fact not only are we going to

use it we're going to use it to good

effect and here's are just some examples

I got three images here people think of

images like X-rays and you look at them

you make a diagnos actually images are

like

encyclopedias look at the image up there

on the left you saw a brain which was

moving around now you're seeing the

fibers between different parts of the

brain this is one brain that's been

looked in multiple Dimensions 100,000

little picture elements you could Click

on each one of those picture elements

and potentially get all the world's

literature about what happens there what

goes wrong there what diseases affect

this place why aren't we doing that

we're doing it in

meteorology so that you could get out

your iPhone and see if it's going to

rain without looking out the

window you they do it in Aeronautics in

building

aircraft they do it in astronomy to

discover How the Universe expanded but

in medicine

trivial take up on the right there you

see a lovely little picture of nerve

cells in the rat brain each nerve cell

carries a function and at the bottom you

see a really nice movie because this one

tells you in picture form how a rat

brain evolves into a human brain over

many many many

Millennia so pictures are there because

they Federate data they integrate it and

they convey information look at this we

we we got a lot of information from

researchers in France and then from a

big website in the States called the

Alzheimer's disease neuroimaging

initiative they just put their data on

the web and said the scientist use them

and we have used them and we found lots

of different bits of

data we found clinical data we found

scans we found proteins we found gen

they gave us all this stuff and we found

a way of integrating them together and

seeing whether they form

patterns and what you see here in blue

dots are patterns of different types of

normal people between the ages of 50 and

90 and in pink you see patterns of

different types of

dementia you might think well why

different types of normal people well

the brain is an extraordinary thing you

have to lose a lot of brain to lose a

cognitive function in Parkinson's

disease you have to lose 70% of a

particular set of brains to start

beginning to show a bit of a tremble

there's that sort of Reserve in

there and with Alzheimer's disease for

example you can see postmortem brains

which have got quite a lot of

Alzheimer's disease in it already but

the people think and act and behave

normally so there's compensation so

firstly you would imagine there would be

normal AED people without any

Alzheimer's and normal AED people who

have compensated our times and we're

beginning to find them and the little

black dots around them are different

types of genes and then there's another

color of dot around the yellow ones

which are different patterns of loss of

brain which you see on scans and it's

the computer that has pulled out of all

the data which we've Federated first

then integrated these different

patterns and in fact we can diagnose

Alzheimer's disease Now with an accuracy

of 95 % as compared to a full clinical

work up in a hospital of only

70%

so these patterns are the important

things we call them disease signatures

and with these disease signatures we

want

to completely transform our diagnostic

catalogs diagnosis should not just be

based on what the patient said and what

you found when examp but should be based

on all the other information as well so

this is bringing in biology to the art

of medicine in order to create disease

signatures to allow us to visualize the

data to allow us to test hypothesis to

see whether something we think about the

disease is correct or not using vast

amounts of data at the moment in the

medical literature the big issues are we

can't reproduce our results

why not because we have small little

groups of

patients but every hospital has small

little groups of patients put them all

together and then you really get the big

picture that's the idea behind it

all so what we want to do is to move

from down

here from the little molecules of DNA

which make up our genes through the

proteins to the cells to the way they

link up

working out the rules which tell us how

these things are structured and how they

limit the way the next level can be

structured all the way up to cognition

and love and memory and feeling and

seeing and talking and understanding so

that when we're faced with patients who

have abnormalities of these things we

know which areas of the brain we should

be targeting where we should focus our

inquiry how we can work out what sort of

prognosis people with a particular

disease signature

had

so this sort of investigation and

examination of data about for example

dementias allows us to put the different

disease signatures together they're the

little pink

balls and you could see that some of

these pink balls appear to be related to

other pink balls around

them through the data that we've been

able to acquire others form like little

satellites all on their own they clearly

have a different

mechanism and we can't just tell that by

ordinary doctr we've got to use

Computing and just to really Hammer the

point

home we also have some proof that these

different disease signatures mean

something because some of them tell you

that you're going to lose your mind the

blue line over a course of about eight

years those are months at the bottom

whereas if you have a different pattern

then you don't lose your mind in the

same way or you lose it much more

slowly so this is real prognosis and

this is also prognosis of subgroups so

that when you apply your treatments you

know whether you're really having effect

or not having an effect now of course

these sorts of

things worry people because many people

still think of their doctor as part

nurse part person who has good taste and

uh

therefore puts good Furniture into their

waiting room and runs a good appointment

system so you got to you got to get away

from that you've got to move forward to

someone who's more technically minded

who who also has a human aspect there no

reason why should or she should lose

that but may have colleagues who might

be doing that side of things but there's

got to be a group of doctors who are in

there trying to find out what is

happening in the brain when it ages

when it goes mad when it becomes

demented and we're facing issues in this

program most of which we've now sorted

out the main issues are the ethical

issues of privacy so our strategy is not

to put data together in big warehouses

and use up all the energy that that

requires and all the space that requires

and all the people to run that that

requires

we've developed a new sort of software

we're going to put at each hospital so

that we would

distribute our questions to all the

hospitals rather like you do when you're

on your computer in your Google so that

our questions will go out and look for

the information they require in the

hospitals and the hospitals will remain

the same and will remain Proprietors of

their data and then we will bring back

just that information so we'll never

bring back any information about an

individual to give us the

answers now if that gives us the sorts

of insights we're now getting you as a

patient will go to your doctor you say

can you please use that information

these guys are sorting out to see

whether you're right to see whether

there might be other things that could

help

me make your doctor embrace the

informatics Revolution make medicine

embrace the infatic resolu uh Revolution

because if we don't have that then we

will remain roughly in the state we are

and the tsunami of dementia and of Psych

psychatric disease in the young with all

the suicides and even murders that come

from

that

untouched that will be our

responsibility the other thing is issues

of consent and ethics and so on most

people like you like me are really quite

happy about the fact there are ethics

committees out there that have been

really very good at managing research

done as it currently is done where you

use a lot of data about

individuals so in all the sociological

surveys when people say who do you trust

in the population doctors are always

first it's a bit silly really but

doctors are always first so the

population doctors have a special

relationship if we can build on that

build on the value and credibility of

the science you patients will be driving

the Revolution and you really really

need to do that you need to say to

someone who when you go to a hospital or

go to doctor surgery says can I use

these data for research purposes you

need to be comfortable to say yes of

course as long as it's for medicine and

not for nuclear

war right you need to be able to say

that it's your data after all you paid

for it but it's also a little bit my

data because I also paid for it through

my taxes and public health is one of our

rights privacy is another right

we have to balance our rights and at

some stage we have to be very careful

how we do that balancing but balance we

must and choose we must we have the

technology now to do all this we've put

it together into something called the

medical informatics platform we're going

to now develop it over the next two

years so that we can make a network of

six hospitals in Tel Aviv in lisan in

Paris in friborg in in Germany uh

potentially in London

and in

Milan we're going to show that it works

and then we're going to put it out to

Tender and get some European company to

start putting it into our hospitals and

start raising the game thank you very

much for

listening