Planning a BioMedCEP project proposal

[rainer93138]

Diabetes - prediction based tool to optimise the therapy

This year the EU project DIAdvisor™ will be finalized which would fit with our BioMedCEP-, HET- or Wearables/Self-Tracking threads of this forum. The consortium is working on a prediction based tool to optimise the therapy of Diabetes.

This project could be enhanced or continued by U-CEP respectively by the use case as suggested in our BioMedCEP proposal ...
More: viewtopic.php?f=13&t=261&p=1335#p1335

[rainer93138]

Yesterday we decided in our core-team to document the worthwhile dialogue of the last days in this forum and to make it visible for all who are interested. Because especially to submit an IP to an open or future call is "a bit" political, we must discuss and decide carefully to avoid frustration and a waste of time and effort. We are working on a strong consortium and are inviting and merging with other related communities:

We wrote to the scene of Neuroscience as discussed, e.g. to Wolfgang Maass/TU Graz, involved in Brain research projects or initiatives (BrainScaleS, Blue Brain, FIAS, etc.) since more than 10 years:

"... we are working on a project proposal, BioMedCEP, draft
viewtopic.php?f=13&t=316&start=50#p1330

We'll submit a pre-check within the next weeks
http://www.citt-online.com/downloads/Bi ... rop_en.doc.

Your research in the field of neuroscience and AI as well as your participation in BrainScaleS fits very well with our proposal (see some pictures/slides 4-6
http://www.citt-online.de/downloads/WP- ... MedCEP.ppt)..."

We also contacted e.g. John-Dylan Haynes/Max Planck Institut für Kongnitions- und Neurowissenschaften Leipzig (Brain Computer Interface / Mind Reading), MPI für Biophysics, Hartmut Michel ("optogenetics based combination switch" as a potential event adapter for living cells) and MPI für Psychiatrie, Florian Holsboer (use case Depression).

There are of course a lot of people from BCI community who could be asked as well, e.g.:
http://future-bnci.org/index.php?option ... Itemid=101
Femke Nijboer is also addressing the problem (as we do), that we must work on getting more Big Dogs but also Small Cats into the scene
viewtopic.php?f=13&t=295

[rainer93138]

Am 13.01.2012 12:45, schrieb Wolfgang Maass/TU Graz:

... I like very much the Science-fiction flavor of this proposal idea, which I did not encounter in previous EU-projects where we participated. It would be fun to see, whether some of our theories for spike based processing and learning could also be fruitful in this expanded context.

Unfortunately I am already participating in another IP proposal for the NBIS call in April. Since the funds for this call are rather low, there will probably be only one IP be funded. Because of this direct competition, I would find it improper to participate in two proposals (and I also have to say, that the proposal in which I am participating fits really well into this NBIS call).

On the other hand, you might actually consider to submit your proposal to another call (with the same deadline), in the Cognitive Systems and Robotics program. Target outcome b, see p.39 of the EU workprogramme,
http://cordis.europa.eu/fp7/ict/docs/3_ ... ict_en.pdf

seems to fit especially well to your proposal idea. This call has also substantially more funds to give away. I understand, that inspiration from neurobiology is an option there, but not required.

I would be concerned about the chances of your proposal in the NBIS call, not only because of the small budget, but also because it will be in my view be quite hard to make the link to neurobiology credible and scientific fruitful. The problem is, that it is still pretty much unknown, what and how neural circuits compute (although we are working on it, and small progress emerges). Hence it is hard to draw inspiration from existing results. Also, the most successful proposals in the type of call of NBIS (like SECO, Brainscales) seem to stay much closer to neurobiology...

[rainer93138]

Am 13.01.2012 12:47, schrieb Rainer von Ammon:

Dear all,

please read this answer of Wolfgang Maass carefully... Wolfgang is involved in BlueBrain/EPFL and BrainScaleS etc.

Before I call him, please tell me your opinion to reshape our proposal again and to go for the mentioned objective

Cognitive Systems and Robotics program. Target outcome b, see p.39 of the EU workprogramme,
http://cordis.europa.eu/fp7/ict/docs/3_ ... ict_en.pdf

As I wrote some weeks ago, it was also my concern that the budget of the call 9/NBIS might already be seen for proposals of the BlueBrain or Quantum community which is working on the Human Brain FET Flagship proposal.

[rainer93138]

Am 13.01.2012 16:49, schrieb Andrew Hunter:

It sounds like a very well-informed heads-up, so I would suggest heeding the advice...

[rainer93138]

Am 13.01.2012 18:27, schrieb Erwin Vlugt/TU Delft:

I also believe not to go for NBIS. As you know, I was skeptic from the start. I need to think about changes plans to the Robotic call. Let you know soon...

[rainer93138]

Am 14.01.2012 10:18, schrieb Dimitris Iakovidis:

... The feedback we received from Wolfgang is important and it seems that we will have to change plans. The robotic call seems relevant, especially with what is described in target outcome (b).

The BioMedCEP idea is "high risk" and compatible with the STREP objective, considering BCI as a component e.g. of an exoskeleton. It could also be compatible with an IP considering that we would like to develop a system "expected to encompass all stages of the research and development lifecycle, coping with cutting across research topics". However, its a bit unclear whether the target "system" should be a robotic system or is it enough to be a system such as the one we would like to propose. May be a communication with the EC would help in resolving this issue...

[rainer93138]

On Sat, Jan 14, 2012 at 1:30 PM, Rainer von Ammon wrote:

First a summary of the suggested next steps:

1) Telco Wednesday, 18 Jan with Wolfgang Maass in order to start a merger between the Blue Brain/Human Brain Project (HBP) / BrainScaleS and our BioMedCEP communities (VPH-FET, U-CEP). If somebody likes to join, Wolfgang offered the afternoon, 4 pm CET (skype, so I would need your skype address)
2) Telco 19 Jan with John-Dylan Haynes/MPI Neuro and Cognition Science/Leipzig and the other addressed MPIs Psychiatrie and BioPhysics
3) Submission to NBIS pre-check after 19 Jan if consortium including MPIs is nearly stable
4) Submission to Challenge 2: Cognitive Systems and Robotics, Objective 2.1: Cognitive Systems and Robotics, b) Cognition and control in complex systems
"Robotics" does not exist in Call9 anymore and we might fit very well with "Cognitive Systems":

"Enabling technologies based on the
acquisition and application of cognitive capabilities (e.g., establishing patterns in sensor data,
classification, conceptualisation, reasoning, planning) for enhancing the performance and
manageability of complex multi-component and multi-degree-of-freedom artificial systems,
also building on synergies between cognitive systems and systems control engineering. This
outcome complements Objective 3.3 / target outcome (d).
Realistic, highly demanding, scalable real-world scenarios will motivate and guide research
related to targets a) & b), and serve to validate its results. Specific Targeted Research Projects
(STREP) are particularly suited to high-risk endeavours, breaking new grounds, with high
potential rewards. They are also appropriate for component-level research for particular
domains. Integrated Projects (IP) are preferred for system-oriented efforts; they are expected
to encompass all stages of the research and development lifecycle and, where appropriate,
cutting across research topics."

If you like to read some more background, this is what I discussed with Bernard de Bono yesterday:

1) NBIS shaphing of BioMedCEP
(...) NBIS actually seems to be an initiative of the Blue Brain/Human Brain Project (FET Flagship 2020 and Beyond proposal of Henry Markram led group) / BrainScaleS (U Heidelberg Karl-Heinz Meier+ EPFL et al.). Wolfgang Maass works with them since around 10 years and he is involved in the IP consortium which is already prepared very specifically along with the NBIS call (...) (see our discussion viewtopic.php?f=13&t=316&start=30#p1292, our community would have to work with the EC from the beginning when the Calls and their contents are defined, we cannot simply submit an IP proposal normally).

2) BioMedCEP NBIS pre-check
In order to get awareness of the EC and the evaluator scene, we should submit the one-pager with a stable list of consortium members.
The result would be interesting and helpful for shaping our future work nonetheless and how to fit in Horizon 2020 (FP8) and also in a FET Flagship. We believe that one of the flagships which will be started in 2013 for 10 years/1 billion € might be a merger of HBP and Quantum Computing. We already tried to shape BioMedCEP accordingly to merge U-CEP and VPH-FET with them. In so far, the pre-check would make sense (...)

3) What to do if the pre-check is positive?
There are also serious critics (...) (e.g. CalTech Christoph Koch at Humanity+ conference 2011) and e.g. an open letter from Zurich regarding Blue Brain/HBP which eats all the bugdets of the EC (...) Insofar our proposal is working on concrete (clinical) use cases and also on concrete product ideas (bidirectional Exocortex/Human Enhancement Technologies in combination with BCI's, Wearable Technologies, Body Area Networks, Event Adapters for living cells, Cognitive Chip technology, etc.).

So, it might be that the evaluators judge positively (...) We believe that the FET team would tell us our real chance with the pre-check.

(...) agreed to wait for the answers of the Max Planck Institutes until end of next week and, if so, we would have a strong consortium. And we would have tried to merge with the HBP/BrainScaleS community what is wanted by the FET team (...) in connection with the flagship proposals (...)

4) Check the submission to 6.2 Challenge 2, objective 2.1 b
We would avoid competition with HBP/BrainScaleS (...) Although the scene seems to be very much broader, we should first check our chance with the head of unit/PO (...)

5) Continue with the BioMedCEP focus anyway
We agreed to continue and not to stop now (...). Bernhard means that what we are doing already, is worth our time and effort and we have already learnt a lot by merging our ideas (...)

[rainer93138]

Am 15.01.2012 14:17, schrieb Andrew Hunter:

(...) definitely a merger is desirable if this can be achieved - there would be good synergies and we avoid a competition we probably lose.

Regarding cognitive systems and robotics, there are huge issues over assistive care and robotic acceptability, and also over how robotics can be developed to be genuinely useful and *safe* in assistive scenarios. Everybody in the assistive care community knows this.

Pitching at cognitive systems that *enhance* the effectiveness of the human while *protecting autonomy* (rather than introducing a separate cognitive agent) seems to me (with a long background in assistive care) a really innovative and potentially game changing approach. The cognitive systems call is really focused on autonomous cognitive systems, whereas we would be looking at an augmentative system with a strong application area focus. If the EU are interested in that potentially very different approach then we have a great opportunity - but you want their confirmation they think this fits with the call.

[rainer93138]

Am 15.01.2012 17:07, schrieb C.G.M.Meskers@lumc.nl:

From a (neuro) physiological/medical point of view there are also good arguments to embrace an approach that challenges and stimulates humans to function "on their own steam" instead of developing technology that may induce passivity (lean on the machine).
This would lead to a proposal that is obvioulsy driven by a strong vision.