In February 2025, a significant joint workshop organised by three major leaders in Brain-Machine Interfaces will take place in Gardanne, France. This workshop is a setellite event of the Braincoder conference.
The agenda will include speakers among the B-CRATOS partners and experts from the Institut Pasteur, the CEA-Clinatec and the EPFL.
More information in the event page
You can register here!
B-CRATOS is excited to announce the publication of the paper Accurate neural control of a hand prosthesis by posture-related activity in the primate grasping circuit, co-written principally by our partner Andres Agudelo-Toro and Hans Scherberger. This was a multi-year effort to decode the principles of grasping actions in the brain.
Andres, a neuro-engineer at DPZ, worked with primates and collaborators in Canada and Taiwan to achieve this breakthrough towards improving BCIs. This paper – published in Neuron on 16 October 2024 – presented for the first time the neural control of the posture of a multidimensional prosthetic hand, paving the way for future interfaces exploiting this additional information channel.
For further information, please refer to the full article and the press release published by our partner Deutsches Primatenzentrum GmbH!
Grenoble, 07/10/2024
B-CRATOS – Press release
B-CRATOS integration of cutting-edge wireless technologies
|
Figure 1: The B-CRATOS Integration Bench test setup at BRME |
Following several months of intense work, B-CRATOS reached a significant stage for the development of the whole project: the complete wireless closed-loop system on bench. Through the advancement and integration of cutting-edge wireless technologies ̶ the low-power RF Backscatter technology and the FAT-IBC technology ̶ closed-loop functionality was successfully verified in a bench test setting.
Main features & objectives
The features of this integrated system, aimed at demonstrating this closed-loop, are twofold: on the one hand, the system captures neural data from cortical regions to control movement, on the other hand, it delivers movement-related sensory feedback to appropriate cortical regions of the brain through direct electrical stimulation.
The key objectives of this B-CRATOS stage are equally wide-ranging. Besides wanting to demonstrate the closed-loop control, here the system integration evaluated performance by measuring latency and data loss.
Regarding latency, the goal is to understand the time it takes for the information to travel from the sensors to the prosthetic and back. Both trips should take less than 100 milliseconds (ms). As for data loss, it is currently measured by counting how many data packets do not arrive properly.
The automatic closed-loop control explained
In the B-CRATOS bench test system, we simulate closed-loop sensorimotor control on the bench. Test data in a neural format, captured by the implantable device, is sent via wireless off-body backscatter and in-body Fat-IBC system to the AI decoder, which processes the data and triggers movement of the prosthetic.
During movement, sensors embedded in the prosthetic hand provide sensory feedback (proprioceptive relating to touch), which is detected by the AI module. A stimulation control command is then returned via wireless pathway to the implantable device.
In this test, when the feedback command is received by the implantable device, the test data is updated to a new pattern, and the loop repeats, demonstrating the continuous function of the full closed-loop system. In a real-life setting, the stimulation would provide the user with feedback relating to the movement and result in a change in subsequent neural motor outputs.
|
Figure 2: The closed-loop data flow for the test setup (clockwise direction). Test Pattern A is transmitted to AI board which results in the Mia hand closing, generating touch and a feedback signal indicating the implant to now generate Data Pattern B. The Loop repeats, with the Mia hand now opening. |
What’s next
In bench tests, the integrated system has successfully demonstrated closed-loop hand control, achieving latencies of under 100 ms for both readout and stimulation pathways. However, the B-CRATOS team aims at optimizing data transmission and reducing latency further, targeting a data rate of over 32 Mbps to support additional electrode channels and/or high sampling rates. These improvements will be informed by the knowledge gained from the current system integration tests and iterative improvements through further refinement and development.
On 16 and 17 October 2024, the B-CRATOS consortium was invited to hold its Consortium Meeting near Lyon, à Romanèche-Thorins. After a warm welcome by Paul Wanda and Cyril Beguet, our colleagues from Blackrock Microsystems Europe and organisers of the whole event, we started our discussions on the project status and the objectives for the coming months.
It was also a very important opportunity for all of us to see the demo system bench, a significant step for the development of B-CRATOS: after several minutes of tweaking and a bit of background stress, BME and NTNU managed to get it working and impress all the partners present!
They were two very intense days that included many debates on the following steps but also the needs of a possible extension of B-CRATOS. All with the wise advice of our Advisory Board members!
Thanks to all the partners present for their contribution and a special thank you to Blackrock for welcoming us warmly and organising the event down to the last detail, letting us discover this region of France and many of its peculiarities!
In a major step forward for next-generation implantable BCI devices, Synchron has just announced positive safety results from their COMMAND clinical trial, reporting no serious adverse events in brain/vasculature of the 6 patients enrolled over 12 months! Impressively, surgeries had a median deployment time of 20 minutes, demonstrating a major advantage of the Stentrode technology. And importantly, the trial demonstrated efficacy, showing that motoric brain activity was captured and transformed to digital control signals.
The Synchron system is a great example of thinking outside of the box to go beyond the current state of the art and enable BCI systems to be potentially safer, longer lasting, and easier to deploy. The future of BCI is arriving and will be made possible by next generation technologies, whether it is novel electrodes like the Stentrode or improvements in wireless implantable technologies, such as those being advanced by our B-CRATOS project, among others.
Find the full article here!
In October 2024, our partners from DPZ participated at the Society for Neuroscience in Chicago. It was a great opportunity for them to share their research and experience with scientists from all over the world.
The poster they presented was Training of real-time robotic grasp decoding from neuronal population activity in macaque motor cortex (M1). It concerned with the primate cortex which can be viewed as a closed-loop dynamical system with very high dimensionality. Extracting a sub-sample of this activity and using it for the decoding of any encoded intention is a challenging task by virtue of the size of the sample relative to the full population. The problem is amplified by the brain’s plasticity, which renders any effort towards decoding ephemeral.
Here they employed a modified Kalman filter to decode the grip of a robotic hand from population spiking activity recorded from two 64 channel Utah arrays implanted in the hand area of primary motor cortex (area M1) of a macaque monkey. Their results confirm that a modified Kalman Filter is suitable to successfully translate cortical activity to motor commands of a robot hand in real time. They also demonstrate the gradual learning by observation and internalization of the task by the monkey using two metrics – the reduction in the transitory nature of the decoder and the speed of the transfer of control of the prosthesis to the monkey. In early experimental sessions, the monkey was slow to take over control and the decoder required retraining within the experiment session, whereas in later sessions decoder training was required only at the beginning of the session and transfer of control was rapid. Furthermore, decoder training involved not only the training of the decoder on the neural activity, but it also adapted brain activity on the decoder, and an equilibrium had to be maintained between the learning processes. This equilibrium was achieved faster and maintained more robustly as the monkey became more proficient in the task, hence demonstrating the viability of this learning-by-observation paradigm.
We are thrilled to share with you a short teaser about B-CRATOS!
Electrode engineers are pushing channel count and density of modern implantable electrodes by orders of magnitude, but today’s BCI systems are unprepared to handle this volume of data. Current day implantable neural interfaces already sacrifice both the quantity and quality of data collected, largely due to constraints on wireless data transmission, due to reliance on technologies such as Bluetooth or inductive coupling. Additionally, users of these devices often risk surgical complications or infection due to the implanted wires and batteries needed to transmit data and provide sufficient power.
The B-CRATOS neural bypass platform technologies look to address these challenges to enable future BCI to unleash the full power of the user’s brain power. Utilizing the novel B-CRATOS in-body and off-body wireless technologies, a BCI system can now capture and transmit BIG DATA neural recordings, effectively leveraging high density electrode technologies and capturing fine-grained changes in neural activity. And as a result, future BCI users will not only be able to “point and click”, but take advantage of the most effective algorithms to control complex prosthetics, decode speech, or meet other everyday needs. Detection routines for closed-loop stimulation devices will not be limited to monitoring biomarkers on a small handful of channels at low sampling rates. And critically, B-CRATOS technologies reduce the need for implanted wires and bulky batteries, which promotes user safety, system longevity, and ultimately adoption of the devices to make a positive impact on human health and quality of life.
BCI Pioneer (and 2024 B-CRATOS Webinar participant) Ian Burkhart has recently co-authored a comprehensive review of clinical trials for implantable BCI in Nature Reviews Bioengineering!
Importantly, Ian and his co-authors not only review the history of BCI trials, but provide valuable insights regarding a pathway forward regarding patient ethics, a need for BCI standards, diversity in recruited volunteers, and other needs/barriers for progress. The truly leave no stone unturned in covering everything from patient participants to research groups to technologies to the practicality of future clinical products.
And given Ian’s unique position as a former implantable BCI study participant, such perspective is invaluable for the field. We expect this review to be a core resource for those wishing to better understand the history, present, and future of the field.
The review nicely compiles a list of emerging electrodes as of December 2023, and one notes the ever-growing number of data channels these devices aim to collect. One can imagine that with multiple such devices implanted, we will see exponential increases in the amount and complexity of data collected in BCI of the future. To take advantage of the full potential of such data, we will need next-generation wireless technologies capable of transmitting it for analysis and storage. At B-CRATOS, we aim to develop such technologies to bring big data BCI systems to impactful real-world applications through improved integration within the body and to devices outside of it. We also expect that by developing next-gen wireless technologies for implantable systems, patient acceptance, usability, and safety can be greatly improved over current systems.
Find the full article here!