Big data in Europe--European Union human Brain Program

The European Union's human brain is scheduled to be selected for the EU's future flagship technology project in 2013, with 1 billion euros in financial support and the world's most important human brain research project, the authors note. At present, there is no need to repeat the importance and significance of the study of human brain, the domestic discussion on artificial intelligence and neuro-cognition is increasing, but it is very simple, but the introduction of the European Union's human Brain program is only a few brief press releases, and the appearance of "Baidu Brain" just attracts the eyeball Therefore, the author wants to provide some practical reference and inspiration to the research institutes, Internet companies and enthusiasts in this field through the detailed description of the EU human Brain program. At the same time, we also want to let you feel, when we are still in the "strongest brain" as entertainment, foreign research on the human brain to how much attention and implementation. Maybe in the future we will still be able to see the familiar picture: we are processing the low-cost brain hardware on the assembly line, while happily talking about how foreign human brain technology is beyond imagination and exciting. Like we talk about computers, the Internet and the iphone today.

Understanding the human brain is one of the most important challenges of the 21st century science field. If we can deal with this well, we can better understand brain principles, develop new treatments for brain diseases, and create a new and revolutionary information and communication technology (information and Communications Technologies, hereinafter referred to as ICT).

This article extracts from the Executive report of HBP (Human brain program, Human Brain Project, hereinafter referred to as HBP), which believes that the integration of ICTs and biology has reached a singular point-it can turn the human brain we have always dreamed of to reality. It is this possibility that prompted researchers to launch HBP's research project (HBP-PS), which is funded by the European Union for one year, with more than 300 experts from neuroscience, medicine and computational science gathering to develop a new ICT to better study the brain and build on it. 2013, the project was selected for the EU's future flagship technology project (Future & Emerging Technologies, FET, another project is graphene), which will receive € 1 billion in funding from the European Commission for 10 years.

I. Background and overview of the project

The researchers found that the biggest obstacle we've ever faced in brain research is the work of brain research and the fragmentation of data produced. Today's research in the field of neuro-cognitive studies is very fruitful but lacks systematicness. It gets a lot of data that describes the different levels of different physiological tissues in many regions of the brain, and these samples come from different species at different stages of development. At present, the researchers urgently need to integrate these data to better show how each part of the combination and form a unified, multi-level research system.

The continuous integration of biology and information and communication technologies has given us the opportunity to achieve this great goal. New gene sequencing and imaging techniques, and new microscopic observation techniques have revolutionized the way we observe the brain. With the Internet and cloud computing, we can easily integrate data from research institutions and medical institutions across the world. Neuro-Informatics provides us with more advanced data analysis methods, which help us to build a very detailed brain map and share, identify our respective knowledge defects and blind spots, and in the absence of experimental data through technical means to predict the value of the parameter. The advent of supercomputers also makes it possible to build and simulate various brain models for the brain's vast and complex physiological details.

These technologies will accelerate our understanding of the brain, as well as inspire us in two other areas: a new approach to prevention and treatment of brain diseases, and a more advanced computing technology that can transform industry, economy and society.

Ii. The new foundation of Brain Research

HBP pursues four goals, each of which is based on existing work and becomes the trigger point for the next study.

1, data

Collect screened, necessary strategic data to map human brains and design human brain models, while attracting research institutions outside the project to contribute data.

Today's neuro-cognitive studies have accumulated vast amounts of experimental data, and a lot of original research has brought about new discoveries. Even so, the vast majority of the core knowledge needed to build a multi-level brain atlas and a unified brain model remains missing. Therefore, the first task of HBP is to collect and describe the filtered and valuable strategic data, rather than the aimless search. Hbp-ps defines three key points for data research:

1 The multi-level structure of the mouse brain. Previous studies have shown that research into mice's brains also applies to all mammals. Therefore, a systematic study of the relationship between different levels of brain tissue in mice will provide a key reference for the Atlas and model of human brain.

2 The multi-level structure of the human brain. The research data of mouse brain can provide important reference for human brain research to some extent, but there is obviously a fundamental difference between them. To define and explain these distinctions, the HBP team should collect strategic data about the human brain and accumulate as much as possible the size of the brain data of the existing mice to facilitate comparison.

3 human brain function and neuron structure. Figuring out the link between brain structure and brain function is one of the important goals of HBP. HBP will focus one-third on the structure of neurons responsible for specific cognitive and behavioral skills, from simple behaviors like other Non-human species to human-specific advanced skills such as language.

2, theory

Define mathematical models that explain the intrinsic relationship between different brain tissue levels and their ability to achieve information acquisition, information description and information storage.

Without a unified and reliable theoretical basis, it is difficult to solve the problem of fragmentation of neuroscience in data and research. Therefore, HBP should include a theoretical research coordinating body devoted to the study of mathematical principles and models that explain the intrinsic relationship between different levels of brain tissue and their ability to achieve information acquisition, information description and information storage. As part of this coordinating body, HBP should establish an open "European Theoretical Neuroscience Institute" (European Cato for Theoretical neuroscience) to attract more outstanding scientists from outside the project, and serves as an incubator for innovative research.

3. ICT Platform

Establish a comprehensive ICT platform system to provide services to neuroscientists, clinical researchers and technology developers to improve research efficiency.

The third goal of HBP is to build a unified technology system that brings together multiple ICT platforms with full technical potential to deal with a new human brain research mission based on ICT. We recommend the formation of six major platforms, neural information systems, human brain simulation systems, medical information systems, high-performance Computing systems, neuro-morphological computing system and neural robotics.

1) Neural information system. HBP's neural information platform will provide neuroscientists with effective technical means to make it easier to analyze the structure and functional data of the human brain and to map out the direction of the multi-level mapping of the human brain. The platform also contains various tools for neural predictive informatics, which helps to analyze and identify statistical patterns of data between different levels of the brain tissue, and also helps to estimate some parameter values that are difficult to derive from natural experiments. In the previous study, the lack of data and knowledge was often an important obstacle to our system's understanding of the brain, and the emergence of these technical tools has made this difficult.

2 human brain Simulation system. HBP will establish a large enough human brain simulation platform to build and simulate multi-level, multidimensional human brain models to address specific problems. The platform will play a central role throughout the project, providing researchers with modeling tools, workflows, and simulators to help them assemble large and diverse data from mouse and human brain models for dynamic simulations. This makes "computer simulation experiments" possible, but not in traditional laboratories that can only carry out natural experiments. Various input values can be generated using various tools on the platform, these input values are essential for medical research in HBP (disease model and drug effect model), neuro-morphological calculations (brain models applied to neuro-morphological hardware), neural robot Research (neural loop models applied to specific cognitive and behavioral tasks).

3 High Performance Computing system. HBP's supercomputing platform will provide sufficient computational power to build and simulate human brain models. It not only has the advanced Exascale secondary supercomputing technology, but also has the brand-new interactive computation and the visualization performance.

4 Medical Information System. HBP's medical information system needs to bring together clinical data from hospital files and private databases (with strict protection of patient information). These features help researchers define "biometric signatures" at all stages of the disease, thus finding key breakthroughs. Once researchers have an objective, biologically based approach to disease detection and classification, they will be more likely to find the root of the disease and develop effective treatment programs accordingly.

5 Neuro-Morphological computing system. HBP's neuro-morphological computing platform will provide researchers and application developers with the hardware and design tools they need to help them develop systems, as well as provide a variety of devices and software prototypes based on brain modeling. With this platform, developers are able to develop a number of compact, low-power devices and systems that are increasingly approaching human intelligence.

6 Neural robot system. HBP's neural robot platform provides researchers with development tools and workflows that allow them to connect sophisticated human brain models to simulated bodies in virtual environments that previously relied on natural experiments of humans and animals to obtain conclusions. The system provides a new research strategy for neuroscientists to help them understand the various multi-level workings of the brain hidden under behavior. From a technical standpoint, the platform will also provide developers with the necessary development tools to help them develop robots that are close to human potential, a goal that could not be achieved without the "brain-like" central controller.

4. Application

The fourth major goal of HBP is to successfully reflect the practical value of basic research, clinical research and technology development for neuroscience.

--A unified knowledge system principle. In this project, "human brain Simulation system" and "neural robot system" will be responsible for specific behavior of the neural circuits, detailed interpretation, researchers can use them to implement specific applications, such as simulation of the impact of genetic defects, analysis of the effects of different levels of brain tissue cell reduction, the establishment of a drug evaluation model. And eventually a model of the human brain that can differentiate humans from animals, for example, can manifest human language abilities. These models will make a qualitative change in our understanding of the brain and can be applied immediately to specific areas of medical and technological development.

--knowledge, diagnosis and treatment of brain diseases. Researchers can make full use of medical information systems, neuro-morphological computing systems and human brain simulations to identify biological signatures in the evolution of various diseases and to conduct in-depth analysis and simulations of these processes, leading to new disease prevention and treatment programmes. This work will fully reflect the practical value of the HBP project. The new diagnosis technology before the disease has not caused irreversible harm, it can be diagnosed in advance, and the actual situation of each patient to develop appropriate drugs and treatment programs to achieve "personalized medical care", which will ultimately benefit patients and reduce medical costs. Better understanding and diagnosis of the disease will also optimize the drug development process, better screening of drug test candidates and clinical test candidates, which is undoubtedly beneficial to improving the success rate of later experiments and reducing the cost of new drug development (the current cost of research and development of each drug is about 1 billion euros).

--Future computing technology. Researchers can use HBP's high-performance computing systems, neuro-morphological computing systems and neural robotics platforms to develop emerging computing technologies and applications. The High-performance computing platform will equip them with supercomputing resources and hybrid technologies that integrate a variety of neural morphological tools. With the help of neural morphological computing system and neural robot platform, the researchers developed a software prototype with great market application potential. These prototypes include home robots, manufacturing robots and service robots that, while seemingly inconspicuous, have powerful technical capabilities, including data mining, maneuvering control, video processing and imaging, and information communication.

Iii. Social Ethics

Considering the enormous impact of HBP's research and technology, the project will form an important social ethics team to finance academic research on the potential social and economic impact of the HBP project, which will influence HBP researchers, manage and elevate their ethical and social responsibilities in an ethical sense, Its primary task is to initiate an active dialogue among stakeholders and social groups with different methodologies and values.

Iv. three major stages of the project

The HBP will last for 10 years and be divided into three important stages.

Initially 2.5 ("ramp-up" phase, ascent), HBP will focus on the creation of an initial version of the ICT platform and collect filtered strategic data for the platform. At the end of this phase, ICT platforms can be used normally by researchers outside the project and project.

The next 4.5 ("operational phase", the operating period) will enhance the collection of strategic data and complement the new features of the platform, and will also actively demonstrate the value of the ICT platform in basic research, medical applications and future computing in neuro-human cognition.

In the last three years ("Sustainability phase", the stability period), the project will continue to work at the last stage and strive to achieve self-financing-making the functions and knowledge created on this platform a permanent asset for scientific research and industrial development in Europe.

V. Cost

It is estimated that HBP's overall investment will be close to € 1.19 billion, with the first phase requiring € 80 million, the second phase requiring € 673 million and the third phase requiring 437 million euros.

Vi. Management and operation of HBP

HBP will be a huge interdisciplinary research project for 10 years, involving partners from more than 20 countries, and is expensive. Therefore, our management mechanism must have strong and flexible leadership, to ensure that the ICT platform truly become a community resources to share and cooperation. HBP cooperation to establish a monitoring mechanism, all kinds of monitoring activities will be throughout the project. Research projects that want access to ICT platforms must undergo a complete competition and screening process, and the process is open to the global scientific community.

Vii. Effects of HBP

HBP will greatly accelerate the overall understanding of human brain structure and function, help human to better study brain disease and find more optimized treatment plan, also can help human develop the revolutionary information communication technology based on human brain mechanism.

From the impact of scientific research, the data collected by HBP and the various technical tools provided by ICTs will help us overcome the problem of the fragmentation of neuroscience research, providing a fresh perspective on the relationship between brain structure and function. The project will give researchers an opportunity to unravel many of the current research challenges in the field of neuroscience, including brain learning and memory, neural coding principles, and even the neural principles of human senses and consciousness.

HBP will also have a major impact on medicine, speeding up the development of new diagnostic tools and treatments. Given the high cost of brain disease, even small advances can have huge economic and social benefits. Reducing drug research and development costs and increasing drug success rates will also benefit the pharmaceutical industry. Core models and technologies developed by European researchers and institutions will greatly enhance the European pharmaceutical industry's competitive advantage in the global brain disease new drug field, which has a huge market potential.

By integrating brain research into the ICT platform, HBP will have sufficient discourse power to determine the future direction of computer technology. Technologies such as supercomputing, human-computer interaction and visualization, omni-directional simulation, and cloud computing, used in projects, will provide more services to industry and consumers, starting a virtuous cycle in which demand growth results in economies of scale and costs, while falling costs further stimulate demand, So that super computer technology can be more widely used in industry and academia. These features require advanced software support, and Europe has a strong competitive advantage in this area.

HBP research in neuro-morphological computation and neuro-Robotics will accelerate the development of low-power systems to reach the human level gradually. While these technologies will not replace the traditional computer technology that drives European development over the past 50 years, their potential applications and strategic significance are equally important. HBP, if able to take the lead in this field, will play a key role in maintaining Europe's competitive position in the world economy.

(editor: Lvguang)