Welcome to the switchBoard official Blog!

SwitchBoard is an In Innovative Training Network (ITN) funded by the European Commission's Horizon 2020 programme under the Marie Curie Actions. The duration of the project is 48 months, starting on November 01, 2015.

The switchBoard consortium brings together eleven beneficiaries from eight different countries, combining the expertise of seven academic partners with excellent research and teaching records, one non-profit research organisation, and three fully integrated private sector partners. This European Training Network (ETN) is supported by six Partner Organisations as well as a management team experienced in multi-site training activities and counselled by a scientifically accomplished advisory board.

Taken together, the switchBoard training network provides an international, interdisciplinary platform to educate young scientists at the interface of neurobiology, information processing and neurotechnology.

Tuesday, 13 June 2017




The main purpose of science is to help people obtain a better understanding of the world and to give us a better future. Researches in neurological field make us not only get a closer look to the delicacy of neurons, more importantly, help human kind to solve neuronal diseases. Scientists have been working on developing prostheses to improve the quality of lives from patients suffering from neurodegenerative diseases. The development of retinal implants, like other prostheses, aim to restore vision for the blind. Here I briefly introduce the principles of retinal prostheses and summarize some representative projects.

Concepts of Retinal Implants

Retinae are composed by very well organized layers of neurons, the photoreceptor layer, the inner nuclear layer and ganglion cell layer (check our other article ‘RETINA: OUR RULES AND CELLS WHO VIOLATE THEM’ for more detail). Most of the retinal implants are designed to benefit patients from retinitis pigmentosa (RP) or age-related macular degeneration (AMD), whose photoreceptor layers degenerate  eventually causing irreversible vision loss. However, most of the patients, even after years of suffering from these diseases, still have the remaining inner cells and ganglion cells in well contact (Weiland et al., 2011). To recover the vision, all we need is to find a good way to compensate the loss of photoreceptors; that is, a device that can sense  light and send the vision signals to the remaining retinal neurons. Therefore, the main task of retinal prostheses is to transform the light signals into electrical signals that retinae can understand.  

Retinal implants are most commonly implemented in three approaches: subretinal, epiretinal and suprachoroidal (Zrenner, 2013).

In the subretinal approach, the implant is placed right between the pigment epithelial layer, which is the layer right next to photoreceptor layer, and the (lost) photoreceptor layer. This kind of implants are usually made by light-sensitive photodiodes that make them able to transfer the light into small currents. They play the role of photoreceptors and rely on the remaining neuronal network for the rest of signal transduction. The advantages of implants from subretinal side are therefore 1. Easy positioning 2. Directly replace the damaged photoreceptors 3. No external cameras are required. 

However, they currently  face the problem of power supply, meaning they either need  huge amount of light in order to generate sufficient current, usually a lot higher than the light from nature environment. Patients today  need to carry an external power source which provides sufficient voltage for stimulation.
On the other hand, epiretinal implants are placed directly on the retinal ganglion cell layer. As retinal ganglion cells act as the output of the visual signal to the brain, implants no longer rely on the remaining neural network; instead, the implant itself directly transfers the images into electrical pulses to the optic nerve. For that reason, epiretinal implants are accompanied by external cameras. The electrical stimuli, compared to subretinal approach, act  directly onto the ganglion cells or their axons and could also help patients even there are barely no remaining healthy cells. Disadvantages are that they are harder to fix on retina since only one side of the implant is attached to the retina, they need an extra force to stabilize the position. More importantly, this approach will need the full understanding of the activities from dozens types of retinal ganglion cells without activating axons of passage.  
In the suprachoroidal approach  the implant is placed between choroid and sclera, and is similar to subretinal approach; however stimulating from a larger distance and therefore requiring larger electrodes. (Luo and da Cruz, 2014).

Few Representative Examples
Argus® II

The Argus® II prosthesis implement a 60 electrodes micro electrode arrays (MEAs) in an epiretinal fashion. Images are captured by a video processing unit adapted to eyeglasses, later on sent to the implant in a wireless way. This implant  has  received a CE mark for medical devices (for Europe) and FDA approval :Currently  more than 100 patients have received these implants.
This implant help patients with bare or no light perception to increase their abilities of recognize and discriminate forms, localize targets, detect motion, and navigate. The best visual acuity is reported by 20/1262.
Each of the Alpha-IMS implants comprises  1500 photosensitive pixels and is implanted in the subretinal side of the fovea, the area with the highest visual acuity. The photodiodes capture light and transfer it into stimulation currents, which activate downstream inner neurons; The external power supply is magnetically attached to a subdermal internal coil (Stingl et al., 2013). This implant has also been commercially available in Europe and is going through human clinical trial. Patients with Alpha IMS implants restore partially the ability of recognition of objects and help them avoid dangerous obstacles on the road. The best reported visual acuity is 20/546.

These are the two most advanced examples of retinal implants. Other ongoing consortia like, Pixium,  Boston Retinal Implant or TSIC, Taiwan Sub-retinal Implant Consortium  are developing own implants.

(Cheng et al., 2017)

Challenges and Future

Despite of the few successful cases and  advances  made over the last decade, there are still challenges for retinal implants.. How are the effects to the chronic stimulation to both the function of the implants and to the remaining retinal neurons remain unclear. The resolution that implants can provided is another important issue. Eeven the best implant so far can only allow patients see objects vaguely. To increase the visual acuity, there are still a lot of engineering challenges to overcome. Other issues like the significant remodeling of neural network after photoreceptor degeneration or the lack of understanding of the interface between retina and the implants are the open questions to be answered.

Although there are difficulties to conquer, simple light sensitivity already help blind people greatly improve their quality of lives. More studies to the fundamental retinal neurosciences are going to help us explore the possibility and to break through the boundaries of technology. In the future, implants with better spatial and temporal resolution can be expected.

Current Status of Projects

(Cheng et al., 2017)


Cheng, D.L., Greenberg, P.B., and Borton, D.A. (2017). Advances in Retinal Prosthetic Research: A Systematic Review of Engineering and Clinical Characteristics of Current Prosthetic Initiatives. Curr Eye Res 42, 334-347.
Luo, Y.H., and da Cruz, L. (2014). A review and update on the current status of retinal prostheses (bionic eye). Br Med Bull 109, 31-44.
Stingl, K., Bartz-Schmidt, K.U., Besch, D., Braun, A., Bruckmann, A., Gekeler, F., Greppmaier, U., Hipp, S., Hoertdoerfer, G., Kernstock, C., et al. (2013). Artificial vision with wirelessly powered subretinal electronic implant alpha-IMS. Proceedings of the Royal Society B-Biological Sciences 280, 20130077.
Weiland, J.D., Cho, A.K., and Humayun, M.S. (2011). Retinal Prostheses: Current Clinical Results and Future Needs. Ophthalmology 118, 2227-2237.
Zrenner, E. (2013). Fighting blindness with microelectronics. Science translational medicine 5, 210ps216-210ps216.


Wednesday, 3 May 2017



Dear SwitchBoard fellows,

Friday 19th of May, Professor Joshua Singer is coming to visit our institute in Innsbruck and he is going to give a lecture with the title "From synapse to circuit: analysis of retinal networks for night vision" at 9 am.
Irem and I are going to live-stream the event and share it with you!
This will give you all the opportunity to listen to his lecture and to ask questions too.
We are going to use Adobe Connect, for the live-conference and the only thing you need to have is flash plugin. You can find the link to the live-streaming here.

For now, save the date!

Lucia and Irem

Sunday, 23 April 2017



In mammals, neural axons (nerves) fail to regenerate after injury. Therefore, you can get paralyzed after bone marrow damage and you can lose your eyesight if the optic nerve suffers a lesion. These are known facts, however, scientists all over the world dedicate their lives to break these rules. Yes the topic is once again breaking the law...and it isn’t just the favourite topic of mine, this is how science reaches new frontiers. We don’t just admit facts and get along with them. For example if doctors say chickenpox is an incurable disease, scientists say ’challenge accepted’, and show the world that in fact, chickenpox is curable.

A research group from the University of California, under the leadership of Andrew D. Huberman, managed to induce optic nerve fibers to grow back to their specific brain regions and therefore to restore some aspects of visual acuity after optic nerve injury in mice. But how was all this possible, if we know, nerves cannot regenerate? Well indeed, the adult central nervous system contains some factors that are unfavorable for axon regrowth, but fortunately, these factors are already identified, which means we can get rid of them.

After optic nerve crush, the axons of retinal ganglion cells fail to grow back beyond the lesion site, eventually resulting in the death of ganglion cells, meaning blindness is irreversible for ever. Huberman and his research group discovered, that downregulation of inhibitors of axon growth, increased activity of intrinsic cell growth-promoting factors and visual stimulation of retinal ganglion cells promote axonal regrowth beyond the lesion site and all the way back to the appropriate brain regions, never missing the correct target. Furthermore, these regenerated axons managed to establish new connections with the appropriate neurons at the target site, meaning the visual pathway was functional again.

These findings were then supported by behavioral tests in which mice showed some restored visual functions (not all of them got restored, for example mice could definitely detect moving objects, however they could not percieve depth) and by specific molecular markers which allowed researchers to trace the regrown axons and verify their target locations.

In summary, with proper stimulation, adult nerves can be induced for long-distance, target specific regrowth and formation of new connections with the appropriate neurons at the taget site thus promoting the restoration of visual function.

By Antonia Stefanov

Saturday, 22 April 2017



I am glad and excited to announce that in May we are going to visit an iniLabs workshop in Zürich, Switzerland. One of our talented early stage researchers, Gemma Taverni - presently completing her PhD at iniLabs -  is organizing a workshop for all the members of the switchBoard project, to introduce the functioning principles and programming opportunities of the neuromorphic Dynamic Vision Sensor. We are looking forward to learn more about Gemma's work!
Please check out the agenda here!

Saturday, 14 January 2017



Human beings naturally build systems and rules around everything they know and the more they know about these things, the more violations they tend to find in the rules. However, according to a legal principle of republican Rome: exceptio probat regulam in casibus non exceptis ("the exception confirms the rule in cases not excepted").

Would you think that we still don't know all the cell types of the retina? Well you should have learnt in secondary school that there are two types of light sensitive cells, called photoreceptors in the retina: the rods and the cones. And then you may still remember that we also have bipolar cells, who connect the photoreceptors to the so called ganglion cells, who then send their long-long axons, which form the optic nerve, to the brain's visual areas, where all the colours and shapes you perceive with your eyes actually become our impression of the world. But those who have learnt about the eye and vision at the university know that things aren't so easy.
Scientists still add new cell types to our list of "known cells of the retina". Well let's put a little more emphasise on the bipolar cells at this point, because they are excellent examples on what I would like to show you here. 

One may think that since we have two types of photoreceptors, we should have two types of bipolar cells as well. That's right: this is exactly how scientists classified these cells at first: rod and cone bipolar cells. These names were meant to tell us a lot about their connectivity – their connections with rods and cones.

Around 2008, researchers thought that they finally identified all the bipolar cell types of the mouse retina. Wait a second, who cares about mice?! Well I have to tell you, that mouse retina is extremely important in research since the majority of the scientists use mouse retina to study similarities with the human tissue. So in 2008 it was believed that there are 10 different types of bipolar cells: 9 types of cone bipolar cells and only 1 type of rod bipolar cell.

As years passed scientists learnt that there still are variations in these 10 types of cells, so they started to create subtypes like type 3a, 3b and so on. At present they think that there are 14 different types of bipolar cells in the mouse retina, out of which only ONE type is a rod bipolar cell, all the rest belongs to the cone bipolar cell "family" (for review, see Euler et al., 2014).

Behrens et al., 2016

Although, that's still not all. The classification turned out to be imperfect since some cells break the rules! Cone bipolar cells, which were believed to contact cones exclusively, were found to have connections with rods (Behrens et al., 2016). Also, surprisingly, the only type of rod bipolar cell also turned out to have connections with some cones. Why is that? Well that's a question many scientists ask themselves worldwide. But trust me, science is evolving quickly, answers are coming soon!

For more information on the switchBoard project visit this site. Have fun, love science!

Thomas Euler, Silke Haverkamp, Timm Schubert & Tom Baden, Retinal bipolar cells: elementary building blocks of vision. Nature Reviews Neuroscience 15507–519 (2014) ; doi:10.1038/nrn3783
      Connectivity map of bipolar cells and photoreceptors in the mouse retina. eLife 2016;5:e20041. doi: 10.7554/eLife.20041 
By Antonia Stefanov