and Its Future
How would you explain The Human Brain Project (HBP) and your role in it to the non-familiar?
The Human Brain Project is one of the two winners of a competition launched by the European Commission as part of its Future and Emerging Technologies (FET) initiative. The goal is to enable breakthroughs in information technology based on understanding the basic organizational principles of the human brain. Over the course of the 10-year project, HBP researchers will simulate the brain’s structure and function — from the single nerve cell up to complex social behavior — and will develop brain-inspired computing technologies. We will also be creating neurobiology-based models of brain diseases.
Our international team of clinicians, computational scientists and statisticians is part of HBP’s Medical Informatics Platform. We are building the technological infrastructure that will allow us to compile data relevant to the brain, which is currently locked in hospital and research archives, without breaching ethical and data protection principles. Rather than hacking into a hospital data information system, we are working on a sophisticated system for distributed queries, which is based on a “Google-like” principle and has access only to local data features extracted behind the hospital’s firewalls. To put it simply, an individual’s health data is processed and averaged within the hospitals system to reach such a level of abstraction that there is no way to (un)intentionally reveal the identity of the given person. We also develop statistical tools that will pave the way for a true revolution in the diagnostics and treatment of brain diseases.
Tell us more about the beginning of the HBP idea: uniting 256 labs frоm 24 countries sounds like a lot of work. How much time did it take to put this idea into practice?
What I can recall is the long preparatory phase that took more than 1 year, maybe even longer. For all of us it was a new playground, even from a managerial and communication standpoint. The challenge to understand the brain is enormous – therefore you need a critical mass of researchers from all disciplines who can master all possible methodological, technological, ethical and even philosophical issues. Sometimes it feels like a big social experiment: weekly virtual meetings across Europe with streaming from London via Lausanne to Tel Aviv and real workshops where you see for the first time the colleagues you’ve been working with. Quite a challenge for a child of the seventies like me.
Currently, HBP is in its ramp-up phase where we are developing six Information and Communications Technology platforms designed for use also outside of the HBP Consortium. The platforms are dedicated to Neuroinformatics, Brain Simulation, High Performance Computing, Medical Informatics, Neuromorphic Computing and Neurorobotics. The idea is to provide open access to the created knowledge through a common web portal. In this way, users who have limited technological expertise will not only be able to access large amounts of data through simple queries, but they will also benefit from the provided analytical tools and be able to suggest improvements.
Do you believe that the mysteries of the human brain will be solved soon, as Henry Markram says?
HBP’s main goal is to revolutionize information technology and clinical neuroscience rather than to create the ultimate Artificial Intelligence product. There are no mysteries to resolve – the real challenge is to integrate the existing knowledge across neuroscience disciplines in a flexible model that accommodates every single piece of new data independent from its granularity level. Translating the physiological processes and architectural principles of the brain into mathematical and statistical frameworks is just a way to formalize the knowledge that we already possess rather than a simplification of “the unexplained”. It can also be seen as a common language that will finally build bridges between the different disciplines – philosophers will talk to physicists, clinicians to computer engineers and so on.
The debate on the opposition between spirit and matter is endless, but HBP is putting the focus back on it. Have you been confronted with the unexplained?
Well, this reminds me of the typical question if the simulated brain will turn out to be male or female, or which language it will start speaking. My position as a scientist is clear –
There is enough accumulated evidence about human behavior and cognition as emerging properties of the brain.
My curiosity about the unexplained is driving my motivation to participate in this project. For me personally, the unexplained is the scarce knowledge about the level of complexity we are dealing with. Therefore we are trying to organize our work according to the principles of brain function — small groups dealing with simple tasks communicate within a network to extract information that is accumulated at higher levels enabling us to move forward. And it works!
It is believed that simulating brain functions will make outcome prognosis easier and will diminish unnecessary treatment expenses. Could incurable diseases be treated one day?
Let’s make an honest reality check. Although we are witnessing an incredible progress in science and information technology, our approach towards the diagnostic and treatment of brain disorders has not changed that much. Why? One of the presumptions is that for significant progress we need collaborative efforts of the whole scientific community – from researchers who have brilliant ideas, but no access to data or equipment, to clinicians who face diagnostic dilemmas and want to benefit from advances in science. If we combine these issues together, the pragmatic answer will be to provide an intuitive platform where existing knowledge can be queried, and limitations in technological expertise are not a burden. This is what our task in the HBP Medical Informatics Platform is about.
Building an accurate model of a given brain disorder at different levels of observation presumes understanding the underlying processes and the ability to simulate and predict аn individual’s outcome. We are still in the process of defining the biological signatures of disease based on objective measures of pathology rather than on subjective impressions of a specialist filtering through his personal experience the information that a patient conveys. This approach did not bring us any further. We still wonder why new drugs fail to provide the expected results, such as treating diseases like Alzheimer’s, and can only shrug when genetic and autopsy results demonstrate that the wrong diagnosis had been made. I very much hope that the HBP will be a game changer and the neuroscience community will follow the good example of the Human Genome Project to achieve the seemingly unreachable in the next decade. This will clearly have a major impact in the field of medicine for the benefit of brain disease patients and their beloved ones.
Could the HBP lead to knowledge that can be used to manipulate people?
Depends on what is meant by “manipulation”. A conventional pharmacological treatment or deep brain stimulation with a pacemaker is also a manipulation, which aims to cure a disease or ameliorate clinical symptoms.
What are the arguments of the project’s critics?
The major criticism from inside the community has been about the management of the project. The way people perceive disruptive business models like Uber or airbnb is similar to the fears of some neuroscientists that this model of disruptive, taboo-breaking science will affect their own research agenda and funding. Human nature.
It is believed that substantial mechanization will occur in the field of medicine in the next 20 years and a lot of young people are not motivated to become doctors as they see no future for the profession. Is there any sense behind that?
I mentioned this in my response to another question. As a clinician, I claim my right to say that
Machines cannot make such decisions, although judging by the recent tragic accident with the Germanwings airplane, humans make errors and their actions can be as unpredictable and dangerous as those of Artificial Intelligence creatures (according to some of the dark future scenarios).
How do smart gadgets and the internet affect the human brain?
This goes beyond my clinical or scientific expertise, particularly on the presumed effects of Internet use on brain development and associated disorders.
I was stunned recently by the number of human years spent on computer games. Smart approaches have started using the enormous creative potential that exists to gain knowledge and test hypotheses. I will illustrate this with some recent developments that have the great potential to revolutionize research through information and communication technology. A British cancer research organization released a free app where Big Data generated through the users in a game-like setting is analyzed to test stochastic models. These help to make sense of a large body of genetic data linked to cancer. Recently, there was this claim against Facebook that they allowed researchers to manipulate feedback on personal profiles to study the specific effects of social network communication on human behavior. Although the approach is ethically questionable, the published data and their interpretation represent a significant contribution to the research in this field. Lastly, neuroscientists claim that computer games in general help to restore 3D vision in amblyopia (the “lazy eye” syndrome) and have long-lasting positive effects on attention and problem solving. The topic remains controversial, but it already breaks the dogma that computers are bad for your (brain) health.
Up to now, I am not aware of rigorous studies on the effects of computer use on social skills and behavior. Let’s not forget that the multi-player online computer games are gaining popularity and also represent a way of socializing – well, a virtual one — which turns a skinny boy into King Kong. But this goes beyond our topic of conversation.
You say that scientists now are miles ahead in what they know in comparison to what they can practically achieve, and also that you don’t see this changing in the next 10 years. What should we expect?
The major issue is the fact that, for some unknown reason, universities and hospitals across the whole world are not investing in state-of-the-art translational research. And this affects all disciplines, not only clinical neurosciences. Indeed, research in a well-equipped laboratory, with fully controlled conditions and trained personnel is a huge investment, but when led appropriately it creates knowledge, and we see the stunning progress made in different domains. The implementation of research-based methods of investigation in the clinical context is always a matter of compromise due to the lack of adequate infrastructure.
If we talk about the future, I am very optimistic that in the next years we will witness a cultural change in university hospitals towards translational research. Given the high level of expectation that society has from politicians and funding authorities, the integration between science and clinics is gaining momentum, and I am convinced there will be major progress in this direction.
Current progress in personalized medicine – treatment targeting a specific disease taking in consideration the individual’s genetic, metabolic, etc. fingerprint is restricted in the majority of cases to a small group of patients. This is mainly due to economic constraints, which pose ethical and moral dilemmas to us as a society.
What do we have to do to keep our brain in great shape?
Avoid stress! Until recently, I was not taking seriously the potential negative effects of psychological stress on the brain. However, in the last few years, a cumulating body of evidence has been coming out suggesting that stress is indeed deleterious, and we should avoid it as much as we can.
Of course, I could jump into my medic’s white coat and start advocating against cigarette smoking, alcohol abuse, etc. Some recent studies suggest that being overweight protects one against dementia, but I am not convinced. This statement could be just one of these mythical inferences generated from overblown studies with negligent sample sizes.
Walk us through a typical day of yours.
The usual day starts at 8 and does not finish before 8 in the evening. Although I try to organize my schedule in a way that allows for a certain minimum of creative work, the truly productive hours are in the evening after everybody has left the building. The amount of emails, meetings and management tasks is overwhelming. Luckily, I have amazing colleagues and friends who make my everyday work as challenging and exciting as it was on my first day.
Our laboratory is organized on a business-like model; we have a number of short meetings — scrums — where we challenge the research fellows, and they challenge us.
How do you relax?
I am a passionate swimmer and could not resist the temptation to join the local masters team. My co-swimmers are an amazing bunch of young, funny and interesting people who share the same passion as me. Our coach is as strict as my coach back in Sofia. Although the days of “we work hard, and we play hard” are over, clubbing in Lausanne is nice and helps to funnel out the negative emotions accumulated through the week.
Skiing and snowboarding are always fun, but my new love is alpine ski touring – away from the slopes, climbing for hours to get the volatile satisfaction of skiing down in just few minutes. Very meditative type of sport, I recommend it to everybody.
Our curiosity poses questions along the lines – why does the human brain, such a complex computational device with virtually unrestricted learning capacity and ability for self-repair, need one banana a day while a supercomputer, not even remotely matching the brain’s function, need a small wind turbines park? Then, how can the gained knowledge be used to better diagnose and treat patients with brain disorders? And finally, how can we make this world a better place to live?