Summary: Researchers developed a compact, low-power brain-machine interface (BMI) called MiBMI, designed to enhance communication for individuals with severe motor impairments. The device translates neural activity into text with 91% accuracy, using a highly miniaturized system that processes data in real-time.
This innovation opens the door to practical, implantable BMIs, promising significant improvements in quality of life for patients with conditions like ALS and spinal cord injuries.
Key Facts:
- MiBMI processes neural signals in real-time, converting thoughts into text with 91% accuracy.
- The deviceโs extreme miniaturization makes it suitable for implantable use, with minimal invasiveness.
- MiBMI could expand beyond handwriting to applications like speech decoding and movement control.
Source: EPFL
Brain-machine interfaces (BMIs) have emerged as a promising solution for restoring communication and control to individuals with severe motor impairments. Traditionally, these systems have been bulky, power-intensive, and limited in their practical applications.ย
Researchers at EPFL have developed the first high-performance,ย Miniaturized Brain-Machine Interface (MiBMI), offeringย an extremely small, low-power, highly accurate,ย and versatile solution.
Published inย the latest issue of theย IEEE Journal of Solid-State Circuitsย and presented at the International Solid-State Circuits Conference,ย the MiBMI not only enhances the efficiency and scalability ofย brain-machine interfacesย but also paves the way for practical,ย fullyย implantable devices.
This technology holds the potential to significantlyย improve the quality of life for patients with conditionsย such asย amyotrophic lateral sclerosis (ALS) and spinal cord injuries.
Theย MiBMI’sย small size and low power are key features, makingย theย systemย suitable for implantable applications. Itsย minimal invasivenessย ensures safetyย andย practicalityย for use in clinical and real-life settings.
It is also a fully integrated system, meaning that the recording and processing are done on two extremely small chips with a total area of 8mm.ย Thisis the latest in a new class of low-power BMIย devicesย developed at Mahsa Shoaranโs Integrated Neurotechnologies Laboratory (INL) at EPFLโs IEM and Neuro X institutes.ย
โMiBMI allows us to convert intricate neural activity into readable text with high accuracy and low power consumption.This advancement brings us closer to practical, implantable solutions that can significantly enhance communication abilities for individuals with severe motor impairments,โ says Shoaran.
Brain-to-text conversion involves decoding neural signals generated when a person imagines writing letters or words. In this process, electrodes implanted in the brain record neural activity associated with the motor actions of handwriting. The MiBMI chipset then processes these signals in real-time, translating the brainโs intended hand movements into corresponding digital text.
This technology allows individuals, especially those with locked-in syndrome and other severe motor impairments, toย communicateย by simply thinking about writing,ย withย the interfaceย converting their thoughtsย into readable text on a screen.
โWhile theย chipย has not yet been integrated into a working BMI, itย hasย processedย dataย from previous liveย recordings, such as those from the Shenoy lab at Stanford, convertingย handwriting activity into textย withย an impressive 91% accuracy,โ says leadย authorย Mohammed Ali Shaeri.
The chip can currently decode up to 31 different characters,ย an achievement unmatched byย anyย other integratedย systems.ย
โWe are confident thatย weย canย decode up to 100 characters,ย butย a handwriting dataset with more characters is not yet available,โย addsย Shaeri.ย
Current BMIs record the data from electrodes implanted in the brain and then send these signals to a separate computer to do theย decoding. The MiBMI chips records the data but also processes the information in real timeโintegrating aย 192-channelย neural recording system with aย 512-channelย neural decoder.
This neurotechnological breakthrough is a feat of extreme miniaturization that combines expertise in integrated circuits, neural engineering, and artificial intelligence.ย This innovation is particularly exciting in the emerging era of neurotech startups in the BMI domain, where integration and miniaturization are key focuses. EPFL’s MiBMI offers promising insights and potential for the future of the field.
To be able to process the massive amount of information picked up by the electrodes on the miniaturized BMI, the researchers had to take a completely different approach to data analysis.
They discovered that the brain activity for each letter, when the patient imagines writing it by hand, contains very specificย markers,ย which the researchers have namedย distinctive neural codes (DNCs).ย Instead ofย processingย thousands of bytes of data for each letter, the microchip only needs to process the DNCs, which are around a hundred bytes.
This makes the systemย fast, accurate,ย and with low-power consumption.ย ย Thisย breakthroughย alsoย allows for faster training times, making learning how to use the BMI easier and more accessible.ย
Collaborations with other teams at EPFLโs Neuro-Xย and IEM Institutes,ย such as with the laboratories of Grรฉgoire Courtine, Silvestro Micera,ย Stรฉphanie Lacour,ย and David Atienzaย promise to create the next generation of integrated BMI systems. Shoaran, Shaeri and their team are exploring various applications for the MiBMI system beyond handwriting recognition.
โWe are collaborating with other research groups to test the system in different contexts, such as speech decoding and movement control. Our goal is to develop a versatile BMI that can be tailored to various neurological disorders, providing a broader range of solutions for patients,โ says Shoaran.
About this BMI and neurotech research news
Author: Michael Mitchell
Source: EPFL
Contact: Michael Mitchell – EPFL
Image: The image is credited to EPFL / Lundi13 – CC-BY-SA 4.0
Original Research: Closed access.
“A 2.46mm2 Miniaturized Brain-Machine Interface (MiBMI) Enabling 31-Class Brain-to-Text Decoding” by Mahsa Shoran et al. IEEE Journal of Solid-State Circuits
Abstract
A 2.46mm2 Miniaturized Brain-Machine Interface (MiBMI) Enabling 31-Class Brain-to-Text Decoding
Recent advancements in brainโmachine interface (BMI) technology offer groundbreaking solutions for individuals with motor impairments, potentially extending to speech synthesis and handwriting assistance. However, current BMIs rely on cumbersome benchtop setups equipped with resource-intensive computing units, restricting their suitability for daily use.
We introduce a miniaturized BMI (MiBMI) system capable of accurate, multiclass neural decoding and high-density sensing in a millimeter-scale silicon footprint, making it suitable for next-generation implantable BMIs. A 512-channel, 31-class neural decoder employs a novel concept of distinctive neural code (DNC) driven by a class saliency model.
This facilitates the precise translation of intricate neural activity into handwritten characters using a low-complexity linear discriminant analysis (LDA) classifier.
The proposed decoder achieves significant improvements in memory utilization (ย โผย 100ย รย ) and computational complexity (ย โผย 320ย รย ) compared to a conventional LDA without DNCs. Moreover, MiBMI enables area-efficient 192-channel neural recording through time-division multiplexing, demonstrating its potential for fully integrated BMIs.
Fabricated in a 65-nm CMOS process, the high-channel-count BMI chipset occupies a compact area of 2.46 mmย 2ย and consumes 883ย ฮผย W. The proposed decoder translated human intracortical neural activity into 31 characters with 91.3% accuracy, significantly enhancing the task complexity compared to previous on-chip BMIs.
Furthermore, MiBMI achieved 87% accuracy in decoding the neural responses of a rat to six classes of acoustic stimuli in an in vivo experiment.
