My thoughts on the standards emerged from TOBI project

I hope that my posts in the last two months provided you a bit more understanding of how the European projects develop, what are the standardization bodies and also made you interested in brain-computer interfaces. At this point I think it's time to move on, looking into other projects and applications about BCI.

Before that, I would like to share you my personal impressions on the standards I have presented so far, during these months. First of all, I admit that maybe the standards TiC, TiA and TiD are not the most representative for how standardization works in EU, since this project was not dedicated to create standards. Even though the standardization was somehow secondary and not what I expected in the beginning, I still find TOBI to be a very interesting project, with great contributions in the area of BCI.

I still have some questions that remain unanswered, due to the fact that most of the information that I have found about this project comes from its official website. Therefore, obviously there are not so many details about the process behind the standardization, focusing more on presenting the work they have achieved. During my research I have even tried to have some insight from the contact persons, but unfortunately I could not get an answer up to this point. One of the main things I wanted to know was if standardization through the official bodies of EU (such as CENELEC or ETSI) was taken into consideration at any point or if the interfaces were meant to be open source files from the beginning.

Regarding the future for this standards, my opinion is that they will not become de facto standards ( as a result of everyone adopting them), but they will help with the creation of other standards, including some parts of them. One aspects that concerns me about their success is that I could not find many references to them except the ones that come from TOBI and that information on them dropped significantly after the project ended.

I hope in my future posts I will refer to this project and its tools as a starting point for other applications in BCI. From this point the blog will focus more on applications and less on the standardization process.

Applications (2)

If last time we saw two applications from Communication and Entertainment areas, today I want to show you other two applications that focus more on the need of person with motion disabilities.

Controlling an Artificial Arm with BCI
Even though the video is in German and probably, as me, not many will understand it, I want to share it with you because I consider it a great achievement in BCI, controlling a prosthesis with brain's activity.



Motor Recovery 
The following video explains how electrical stimulation helps with the recovery of motor sensors.

Once again, more detailed information on these applications can be found here.

Applications

Today I have checked the stats for this blog and I have to say I am nicely impressed not only by the number of visitors, but also by the geographic distribution of the visitors. That tells me I am not alone on this and once more shows that BCI present a great interest worldwide.

However, because I want to attract even more people in this discussion I will show you some of my favorite applications from the TOBI project, hoping that would convince more people about the potential of this area.

Web browsing Assistive Technology
As you can already guess from its name, this tool allows you to surf on the internet while giving BCI-actuated commands. This could be the way the future looks like, replacing keyboards and mouse.



The Brain Painting Application
A participant is painting by concentrating on specific commands.


More applications and detailed explanation can be found here.

Workshops

With such a vast project and partners from so many different places it is obviously very difficult to manage the project. The workshops come in handy on this matter and they provide insights and directions to follow. During this workshops the ideas that are presented by the speakers or the work submitted by the persons who have an interest in the topic are analyzed and filtered not only by the members, but also by all the experts in this domain.

The first workshop of TOBI took place in Graz (Austria) in February 2010 and was entitled "Integrating Brain-Computer Interfaces with Conventional Assistive Technology". The main idea behind this international meeting is, like its title states, to inspect how BCI can be better integrated with the existing assistive technology. Even though there was a huge increase in interest and innovations in the BCI area, the outcome failed to reach the target market (the users) because of the lack in interoperabilty.

"Translational Issues in BCI Development: User Needs, Ethics, and Technology Transfer" was the second workshop in a series of four and took place in Rome (Italy) in December 2010. The main goal was to find ways to reduce the significant gap between the advances in computing capabilities and what neural engineering was offering at that point. In order to so, further improvements were considered, by user centered research and design, neuroethics and technology transfer.

For the third workshop the partners and associates met in Wurzburg (Germany), in March 2012 under the title: "Bringing BCIs to End-Users: Facing the Challenge . Evaluation, User Perspectives, User Needs, and Ethical Questions". While still aiming for the goals presented in the first two workshop, in this meeting the relation Technology-Users was the main focus. By informing the interested people on what the existing BCI technology can offer and listening to their opinions and needs the expectations were that clear directions to follow will emerge.

In the final meeting, which was in Sion (Switzerland), in January 2013 some conclusions on the current state of the progress in BCI technology were drawn. Under the name "Practical Brain-Computer Interfaces for End-Users: Progress and Challenges", this workshop brought together developers, users, professionals, clinicians with the goal of evaluating the current position and future scenarios. Along with this meeting, the project was concluded.


Therefore, you can see that the workshops are a great tool in order to manage a project a project of this size and to organize all the partners. As far as I understand, this is the common practice among European projects of this type.

More detailed information on the workshops can be found here.

Activity of the consortia

So far we have seen that the goal of TOBI project is to bring valuable contribution to the BCI area, by providing tools meant to help the users and companies and future innovations. During my research, I could not find any clear information of how this entire process was structured, how they came up with different subjects and if there was a voting system or not. However, I have found that four main directions were identified and during the period of the subjects the partners and associates gather for four workshops.

The four main applications areas where TOBI focused in order to help people were:

• Communication & Control - just imagine you can control a computer just by thinking of certain actions. This tools are meant to increase the capacity of changing information between user and interface with the help of EEG signals.
• Motor substitution - being able to regain some of mobility and manipulation functions would have been looked as a miracle not so many years ago. With the help of neuroprostheses this is starting to be more and more viable.
• Entertainment - even though it is not nearly as important as the last two, this area also has a great potential. In day to day life we interact with the environment through auxiliar interfaces such as switches, joysticks, computer screens. Being able to express our feelings and preferences just by thinking about it would be a huge step of progress (in fact it already is). The standards and tools that came out of this project could be some of the pieces that glue together future technologies in BCI.
• Motor recovery - as we have seen usually BCI technologies are looked as a replacement tool for the people with mobility disabilities. However, they can also help patients with the recovery. Mental practice of motor actions can regain sensors abilities.

From the start of the project and to its finished a significant numbers of papers, book chapters and books have been published by the partners involved in TOBI. Most of them relate or describe the work developed inside the project. You can find a list of this publications at this address.

As far as my research went all of the work is what we call open source. This means all of these tools can be just downloaded and integrated in projects by anyone who has any interest in them. This is also the case for the standards I have mentioned in my previous posts: TiA, TiC, TiD. They can be found and copied at this address. This further strengths the voluntary aspect of this work.

Standardization in a consortia

Up to this point, one can already sense that the process of standardization in a consortia involves all the partners involved and differs from the one in a formal organization body.

In order to understand the concept of a consortia I have tried to find a definition for it. Unfortunately, I have found that for this type of organization it is very difficult to have a fixed definition since it strongly depends on the area that they are working on and how the consortia came together. However, in this paper I found a reference to a characterization of consortia, as seen by de Vrie in his book on Standardization, published in 1999. Since I consider that these characteristic nicely summarize what a consortia is and its purposes I will share them with you:

• Organizations that do not develop standards themselves, but in one-way or another perform standardization-related activities in relation to SDOs (standards developing organizations).
• Specialised standardization organisations such as industry sector groups, whose purpose is to develop standards. The membership is sometimes open to all interested corporations, universities and governmental agencies.

•  A form of co-operation between competitors to agree on standards: a broad grouping of different companies pursuing a common objective, usually in a particular technology field. 

•  Organizations that companies co-operate in R&D to share cost and, as part of the project, agree on standards.

I consider that the TOBI project is fairly described by the first and last characteristics, since its purpose is not the standardization process itself, but all means of providing useful tools in BCI. However, this is just a personal opinion and I could not find any details on how did the standards presented in my previous post emerged and what kind of mechanism was inside the consortia. The information available on the project does not provide clear information on the process, such as if there was a voting process or not, if all partners had an equal vote or how they dealt with negative votes (if that was the case).

Regarding the standardization regime of TOBI I think it is safe to say that it was a voluntary standardization, open and consensus like we have seen that is the case in these type of European projects.  Moreover, the consortia seems to be market orientated with no government influence at the national level of any of the partners.


References:

• M. Kajiura, A change by the Consortium in ICT Standardization, World Academy of Science, Engineering and Technology, 2008
• H. de Vries, Standardization, Kluwer Academic Publishers, 1999

Overview of the standards described in the TOBI project

If you ever had a course on Standardization or did some research by yourself, probably the case of the Great Baltimore Fire rings a bell. The compatibility is always a great challenge when different systems with different requirements and outputs are combined. That is why through standards a great effort is made to increase the interoperability between systems. In order to do so it is often recommended to have standard interfaces that will adapt to different conditions, in different applications.

Unfortunately, I found that the description for each one of these standards is often vague and only one paper appears to be published describing TiA interface. However, short descriptions of each of the standards are available on the websites related to the project. Below you can find the information exactly how it is provided by the partners.

TiA - Data acquisition interface 
"With TiA we attempted to standardize the first interface shown in Mason and Birch's model between "Amplifier" and "Feature Extractor'". BCI as well as neuroscience research in general would benefit from this interface, as the used hardware becomes irrelevant and easily exchangeable as a result of a standardized abstraction layer. With TiA it is possible to reliably transmit data (e.g., for recording) using TCP or use unreliable transmission via UDP if needed based on the users choice. TiA is separated into two parts: (i) a meta information transmission (e.g., channel names, sampling rates) and (ii) the actual data transmission, transmitting the acquired data to the client(s) in an efficient way with minimal overhead also supporting multirate signals and different signal types (e.g. EEG, EOG, EMG, ... ). at the same time."

TiC - BCI output interface
"TiC defines the way classifiers in the BCI transmit their outputs to other modules. Thinking in terms of Mason and Birch’s BCI model, TiC is designed to provide a generic interface between the BCI Transducer and the Control Interface. It is based on XML and the usual way of transmission is network based (TCP/IP or UDP). It requires little setup. Also, TiC supports a hybrid setting, where multiple classifiers may be transmitting at any one time."

TiD - Markers and events interface
"TiD is designed for flexibility, whilst preserving compatibility with legacy implementations. What this means conceptually is that TiD can work in asynchronous mode, allowing the immediate distribution of markers to all modules that have registered to receive them. In the case of a synchronous BCI, as is the case for most current implementation, it simply means that a single module registers for getting the markers."

As it can be seen all three different standards are related to compatibility.  Because we are more interested in the stages behind the standardization than the technical details of each standard, I will stick to this descriptions. However, the following block diagram should provide an idea of where we can find this interfaces in a BCI system.

The figure was extracted from the paper "Tools for brain-computer interaction: a general concept for a hybrid BCI" and all rights belong to the original authors.

If you are interested in more technical details on the subject, a more detailed presentation can be found here.