Projects

Development of biomimetic neural sensory feedback for neuroprosthetic applications

Peripheral intraneural stimulation can provide tactile information to amputees. However, efforts are still necessary to identify encoding strategy eliciting percepts that are felt as both natural and effective for prosthesis control. Here we want to develop neural modulation strategies able to improve the naturalness and efficacy of stimulation to convey sensory information to trans-femoral amputees implanted with intraneural electrodes.

Keywords: neuroprosthetics, bionics, neuromodulation, somatosensation, neural stimulation, electrodes

Description
Re-establishing the sensory flow of information between prostheses and the brain is of paramount importance. Lack of sensory feedback and inadequate embodiment are among the reasons for rejection of available commercial prosthesis. Implantable peripheral nerve interfaces can be reliably used to provide sensory feedback to limb amputees. This approach can improve prosthesis control, is usable and stable over long periods, and fosters embodiment of the prosthesis by the subjects. Previous works also showed that different encoding strategies could be used to successfully restore sensory feedback. In particular, in these studies the amplitude or frequency of the injected stimuli was modulated, eliciting somatotopic sensations (e.g., referred to phantom extremity) with feelings that are sometimes similar to the natural ones (e.g., pressure or vibration). However, naturalness can be characterized by different ‘‘grades’’ and the ‘‘level’’ of naturalness has been often reported by the subjects as quite limited (quite unpleasant). Ideally, neural stimulation should be able to provide sensory feedback that is functionally effective and highly natural, as the naturalness of the feedback plays a pivotal role in prostheses acceptance. A possible way to address this issue might be to define and use complex stimulation patterns that resemble the natural encoding strategies implemented by the nervous system: the biomimetic neural stimulation strategies. This paradigm should be implemented in a real-time configuration in a bionic prosthesis.

Goal
The student will be guided in understanding the principal causes of lack of sensory feedback, its effects and meaning in terms of artificial perception, the current state of the art of neuroprosthesis with scientific literature readings, and our developed sensory-feedback system. The student will study in detail the mechanisms and principles of direct electrical nerve stimulation for sensory feedback applications. The major goals (mandatory) for the student will be: 1. Processing of neural signals recorded experimentally with neural electrodes implanted in the nervous systems, in response to different neural stimulation. 2. Extraction of key features of biomimicry from different neural signals. 3. Design and implementation of a novel biomimetic sensory encoding based on mechano-neural model of touch (FootSim/TouchSim). 4. Real-time implementation of a biomimetic neuromodulation according to wearable sensors outputs in a closed-loop neuroprosthetic system. 5. Validation of the new system with volunteers/amputees Recommendable skills: Signal processing, MATLAB, C++, C, peripheral nervous system neurophysiology and anatomy. Extra skills: Computational neuroscience, multithread applications, embedded linux system (e.g. raspberry pi).

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Life Bionics, Goteborg Sweden

Emulating skin biomechanics through GelSight: neural mechanisms of tactile texture perception

The project consists in assessing the degree to which our perception of natural texture is shaped by the mechanics of the skin. We have implemented an approach, developed by Ted Adelson at MIT (GelSight), that consists of fabricating a gel whose material properties match those of the skin and then imaging (using a laser profilometer) the pattern of deformation on the surface of the gel that is produced when pressed against the surface. We can then estimate how the skin would be deformed by any given texture using this approach and assess whether we can better predict from these patterns of skin deformation the responses of tactile nerve fibers to that texture and the perception thereof.

Keywords: somatosensation, neural processing, sensory system, touch, neural recording

Description
We showed that the high-frequency components of a texture are preferentially filtered out by the skin. Furthermore, the pattern of skin deformation produced by some textures, particularly textures with highly compliant elements, is almost completely unpredictable from the profile of the texture itself. We are now comparing the predictions of our in-silico model (TouchSim) and Gelsight with nerve recording during realistic texture scans in primates. These preliminary findings are very promising and likely to lead to important insights into tactile texture perception.

Goal
The major goals for the student will be: - Is touch shaped by the biomechanics of the skin? - Frequency components of textures in skin deformation (spatial/frequency filter?) - Does the skin transform textures? - How involved is skin biomechanics in neural encoding patterns? - Compare simulated (TouchSim) neural response with nerve recordings. - How are textures perceived? Predict from the textures the perception. Recommendable skills: Signal processing, MATLAB, C++, C, peripheral nervous system neurophysiology and anatomy. Extra skills: Computational neuroscience.
Time effort required: Master project full time.

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Department of Electrical Eng., Life Bionics, Goteborg Sweden Email: valleg@uchalmers.se

Restoring motor and sensory functions with a wearable sensorized exoskeleton

n this project, we propose a real-time sensorized hand-exoskeleton that combines the exosuit and the sensory feedback based on TENS systems. The integrated robotic system will be able to provide the user with assisted grasping force control via exo and with fine grip-force feedback.

Keywords: bionics, exoskeletons, sensory feedback, neuroprosthetics, bionics

Description
Over 50 million people worldwide suffer from persistent neuromotor hand impairment. We will exploit a robotic hand exo to provide grasp assistance in therapy and activities of daily life. User studies in people with a spinal cord injury (SCI) proved the potential of the robotic hand orthosis but also highlighted room for improvement in terms of functional assistance and provided grasp force. Transcutanoues Electrical Nerve Stimulation (TENS) has the potential to improve grasp function by providing sensory feedback or targeted muscle activation. This project aims to develop and evaluate a proof of concept of the combination of specific patterns of electrical stimulation with the Exo.

Goal
In the first part of the project, the student will familiarize with the exo, gain experience from running user studies, and prepares the required hardware (electronics, actuation system). Based on these insights, the targeted use case will be refined. Afterward, the student will familiarize with the electrical stimulation principles and he/she will adapt the system developed for the specific application. The robotic hand orthosis and electrical stimulation will be integrated to create a closed-loop hand assistive device that improves hand function in people with SCI through the combination of mechanical assistance, augmented sensory feedback, and potentially functional stimulation. Finally, the student will design tasks to evaluate the functionality and benefit of the combined approach in neurologically intact subjects and people with SCI.

Required Tasks 10% Literature review of state-of-the-art 10% Familiarization with the exo and preparation of required hardware 20% Definition of design case and familiarization with electrical stimulation 3 30% Integration of electrical stimulation with the exo 20% Design of functional tasks and assessment of the integrated system 10% Presentations and report

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Dept. Electrical Eng, Life Bionics, Goteborg Sweden email: valleg@chalmers.se

A TOUCH SIGHT: NEURAL BASIS OF OBSERVED AND ACTUAL TOUCH

Touch is a complex, multisensory perceptual process (de Haan and Dijkerman, 2020; de Lafuente and Romo, 2006; Graziano and Gross, 1993). In non-human primates (NHPs), multisensory input (e.g., visual, tactile) converges upon neurons in higher-order brain regions such as the posterior pari- etal cortex (PPC) where they are integrated into coherent representations (Graziano and Gross, 1993; Avillac et al., 2007; Graziano, 1999; Graziano, 2001; Graziano et al., 2000; Holmes and Spence, 2004; Hwang et al., 2014; Seelke et al., 2012; Sereno and Huang, 2014). Recent human neuroimaging studies suggest that the PPC is also recruited during touch cognition in the absence of actual tactile input (e.g., seen touch or imagined touch), supporting a notion that both higher- level touch processing and tactile cognition share a neural substrate (Chan and Baker, 2015; Lucas et al., 2015). To date, however, such a link has not been established at the single neuron level.

Keywords: neuroprosthetics, BCI, neuromodulation, neuroengineering, bionics

Description
Neural Representation of Actual and Observed Touch in Somatosensory Cortex (BA1). Vicarious activation: Activation of a brain region that is normally involved in processing the observer’s own actions and sensations, but that is now activated by seeing similar actions or sensations in another person. In a unique opportunity, we investigate touch processing in a SCI human subject at the level of single neurons recorded from an electrode array implanted in the left S1 for an ongoing brain machine interface (BMI) clinical trial. We have data collected in multiple sessions. We recorded single- and multi-unit neural activity during the presentation of actual touch and during observed touch to sensate dermatomes below the level of the participant’s injury.

Goal
Do SI also contain mirror neurons that are active during both the observation and the execution of actions? Here, we test whether a Broadman’s Area1 neuronal population is similarly engaged in the actual and observed somatosensory domain. The candidate will dig into the cortical neural data to unveil neural mechanisms related to touch. The major goals (mandatory) for the student will be: 1) Is S1 activated when observing someone else hand being touched? Is this activation higher w.r.t. touching objects? 2) Is the somatotopy preserved in vicarious touch? 3) Are the same neurons? 4) Is modality specific (touch-proprio)? 5) Is the activity modulated with observed stimulus properties? Can we decode the stimulus characteristics from the observed touch activity? 6) Could the manipulation of activity in S1 with ICMS change people’s perception of others? (Sensory perturbations) Recommendable skills: Signal processing, MATLAB, central nervous system neurophysiology and anatomy. Extra skills: Computational neuroscience. Time effort required: Master project full time.

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Department of Electrical Engineering, Life Bionics Goteborg Sweden Email: valleg@chalmers.se

Development of a computational model of the human somatosensory nerves for stimulation and recording

Our previous studies has identified that the nature of signals recorded with electrodes implanted in the peripheral nervous system strongly depends on the type of electrode and the degree of activity inside the nerve. For this project your aim will be to further explore which dimension of a population activity a given type of electrode can identify. Based on the literature you will identify discharge patterns combination permitting to recreate the main different classes of bio-plausible population activity. You will also complete the existing hybrid model to integrate the main kinds of electrodes used with neural interfaces (use of Solidworks, Matlab, and Comsol). You will then implement your bio-mimetic population activity in the full model and study each electrode’s dynamic selectivity. Your final goal would be to establish rules permitting to classify from the recording alone what “family” of activity is happening in the nerve.

Keywords: neuroprosthetics, bionics, computational modelling, neuromodulation, neural interfaces

Description
Neuroprostheses based on electrical stimulation are becoming a therapeutic reality, dramatically improving the life of disabled people. They are based on neural interfaces that are designed to create an intimate contact with neural cells. These devices speak the language of electron currents, while the human nervous system uses ionic currents to communicate. A deep understanding of the complex interplay between these currents, during the electrical stimulation, is essential for the development of optimized neuroprostheses. Neural electrodes can have different geometries, placement within the nervous system, and the stimulation protocols (paradigms of use). This high-dimensional problem is not tractable by an empiric, brute-force approach and should be tackled by exact computational models, making use of our accumulated knowledge. In pursuit of this goal, a hybrid finite element method—NEURON modeling—is used for a solution of electrical field generated by stimulation, within the different neural structures having anisotropic conductivity, and a corresponding neural response computation.

Goal
Recommendable Skills: MATLAB, Solidworks and COMSOL, NEURON. Time effort required: Master project full time

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Goteborg Sweden Email: valleg@chalmers.se

 

Perceptual and cortical representations of corporeal touch explored using intracortical microstimulation of primary somatosensory cortex

The somatosensory system is crucial to the formation and maintenance of coherent mental representations of our bodies. Traditional concepts of somatosensation have been shaped by the principles of somatotopic and hierarchical organization of the primary somatosensory cortex (S1). However, emerging psychophysical phenomena have been studied mostly with natural touch only that undergoes extensive processing along the tactile system, and it is unclear at which stages these phenomena arise. Intracortical microstimulation (ICMS) of S1 allows to directly evoke vivid touch sensations on the body, the properties of which can by systematically manipulated by varying the parameters of stimulation. In this work, we use ICMS of human S1 in three implanted participants to define cortical-body maps of the human hand and then to link these mental-body perception maps.

Keywords: neuroprosthetics, BCI, neuromodulation, neuroengineering, bionics

Description
Around 169,000 people in the United States live with tetraplegia due to spinal cord injury (SCI). (National Spinal Cord Injury Statistical Center, Facts and Figures at a Glance. Birmingham, AL: University of Alabama at Birmingham, 2018). The resulting paralysis and accompanying loss of independence cause a severe decline in quality of life and necessitate around the clock care. A promising approach to restore independence to individuals with tetraplegia is to equip them with robotic arms that they can control volitionally via signals harnessed directly from the central nervous system. With the development of ever more sophisticated robotic arms and of interface technologies that yield better control signals, the need for the restoration of somatosensory feedback in Brain-Computer Interfaces (BCIs) has come into clearer focus (Bensmaia and Miller, 2014; Bensmaia et al., 2020; Flesher et al., 2016). Indeed, for able-bodied individuals, interactions with objects are critically dependent on signals from the hand that convey information about the objects and our interactions with them. Without these signals, our ability to interact with objects is severely compromised, as visual signals are poor substitutes for their tactile counterparts. Recent efforts toward developing brain-controlled robotic limbs have thus incorporated artificial sensory feedback by applying intracortical microstimulation (ICMS) to the somatosensory cortex (Flesher et al., 2016, 2021; Salas et al., 2018). ICMS has been shown to evoke stable and nearly natural tactile sensations experienced at specific locations on the (otherwise insensate) hand.

Goal
The student will be guided in understanding the principal causes of lack of sensory feedback, its effects and meaning in terms of artificial perception, the current state of the art of neuroprosthesis with scientific literature readings, and our developed sensory-feedback system. The student will study in detail the mechanisms and principles of direct electrical cortical stimulation for sensory feedback applications. Investigating the functional representations of touch and the relationships between corporeal, cortical and perceptual spaces. The major goals for the student will be: 1. Spatial acuity of ICMS compared to natural touch 2. Relationships between corporeal – cortical – perceptual spaces 3. Granular representation of human fingers in S1 4. Perceptual biases 5. Array design and bionic hand sensorization Recommendable skills: Signal processing, MATLAB, central nervous system neurophysiology and anatomy. Extra skills: Computational neuroscience. Time effort required: Master project full time.

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of the Neural Bionics laboratory, Chalmers University of Technology, Life Bionics, Goteborg Sweden email: valleg@chalmers.se

Biomimetic microstimulation of the human somatosensory cortex

Sensory feedback based on intracortical microstimulation has been shown to improve subjects’ ability to use brain-controlled bionic hands (Flesher et al., 2021). However, the resulting dexterity is still far from that of natural hands in able-bodied individuals. Efforts to sensitize bionic hands for amputees by electrical stimulation of the nerves have shown that sensory feedback that mimics natural tactile signals (so called biomimetic feedback (Okorokova et al., 2018; Saal and Bensmaia, 2015; Saal et al., 2017) evokes more natural and more intuitive sensations that better support interactions with objects than does non-biomimetic feedback. Despite these successes with amputees, biomimetic feedback has never been applied in the context brain-controlled bionic hands.

Keywords: neuroprosthetics, BCI, neuromodulation, neuroengineering, bionics

Description
Around 169,000 people in the United States live with tetraplegia due to spinal cord injury (SCI). (National Spinal Cord Injury Statistical Center, Facts and Figures at a Glance. Birmingham, AL: University of Alabama at Birmingham, 2018). The resulting paralysis and accompanying loss of independence cause a severe decline in quality of life and necessitate around the clock care. A promising approach to restore independence to individuals with tetraplegia is to equip them with robotic arms that they can control volitionally via signals harnessed directly from the central nervous system. With the development of ever more sophisticated robotic arms and of interface technologies that yield better control signals, the need for the restoration of somatosensory feedback in Brain-Computer Interfaces (BCIs) has come into clearer focus (Bensmaia and Miller, 2014; Bensmaia et al., 2020; Flesher et al., 2016). Indeed, for able-bodied individuals, interactions with objects are critically dependent on signals from the hand that convey information about the objects and our interactions with them. Without these signals, our ability to interact with objects is severely compromised, as visual signals are poor substitutes for their tactile counterparts. Recent efforts toward developing brain-controlled robotic limbs have thus incorporated artificial sensory feedback by applying intracortical microstimulation (ICMS) to the somatosensory cortex (Flesher et al., 2016, 2021; Salas et al., 2018). ICMS has been shown to evoke stable and nearly natural tactile sensations experienced at specific locations on the (otherwise insensate) hand.

Goal
The aim of the project is to develop the first biomimetic approach to restore touch via intracortical microstimulation (ICMS). We will test whether bio-inspired patterns of ICMS are perceived as more natural, yield better prosthetic control, and give rise to greater embodiment of the prosthetic limb, as has been shown to be the case with nerve stimulation. The proposed project leverages the fact that our team is one of the few to have two human participants chronically implanted with arrays of electrodes in both motor and somatosensory cortex.
The major goals (mandatory) for the student will be: WP1: develop the first biomimetic ICMS encoding algorithm. WP2: implement the biomimetic encoding algorithm in real-time with the human BCI subjects.
WP3: test the sensory and functional consequences of biomimetic ICMS-based sensory feedback in the human subjects. Recommendable skills: Signal processing, MATLAB, C++, C, peripheral nervous system neurophysiology and anatomy. Extra skills: Computational neuroscience, multithread applications, embedded linux system (e.g. raspberry pi). Time effort required: Master project full time.

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Life Bionics,Goteborg, Sweden Email: valleg@chalmers.se

Calibration Algorithm for Smart Neurostimulation Sock - (collaboration MYNERVA - ETHZ)

MYNERVA is pioneering the development of a smart neurostimulation sock aimed at alleviating symptoms associated with neuropathic diseases of the lower limb, such as those caused by diabetic neuropathy. Through targeted transcutaneous electrical stimulation (TENS) of the nerves innervating the foot, MYNERVA reduces neuropathic pain and restores sensory perception below the foot in affected individuals, improving gait and balance. Empowered by a user-friendly mobile application, patients can independently regulate stimulation in real-time, eliminating the need for professional intervention. Integration of the hardware into a sock ensures seamless integration into patients' daily routines. To ensure adaptability to varying patient anatomies, the integrated arrays of electrodes and pressure sensors within the sock require precise calibration before each use. Conventional calibration methods are excessively time-consuming, rendering daily usage impractical. Engineering a fast and accurate calibration algorithm for both electrodes and pressure sensors is therefore a crucial component in the development of the device.
As an ETH spin-off, MYNERVA has previously developed the described neurostimulation device and validated it through clinical studies.
Diabetic Neuropathy: Patients suffering from Diabetic Peripheral Neuropathy (DPN) often experience severe damage to peripheral nerves, particularly in the lower limb, resulting in chronic pain and sensory loss. Traditional treatments such as opioids carry side effects, while physiotherapy primarily focuses on compensatory walking strategies, neglecting sensory restoration and gait improvement. TENS for Diabetic Neuropathy: Transcutaneous Electrical Nerve Stimulation (TENS) offers promising prospects for addressing chronic pain and sensory loss in patients with DPN. Strategically positioned arrays of electrodes enable precise stimulation of nerve pathways in the foot, which facilitates pain reduction and can elicit somatotopic sensations. By synchronizing the latter with pressure sensor activation, sensory perceptions below the foot can be restored.

Project Description
This master thesis will focus on designing and gathering data for an efficient and robust calibration algorithm. The student will collect and analyze electrode stimulation and pressure sensor data from a group of patients utilizing a prototype device. Various machine learning algorithms will be explored to develop an accurate and fast calibration method for both electrodes and pressure sensors. Subsequently, the resulting algorithms will undergo verification with patients to ensure effectiveness.
Skills: Python, Machine Learning, Clinical Trials.
Duration: Master project (6 months full time).
Contact:
Dr. Greta Preatoni, Project Leader MYNERVA, Email: greta.preatoni@wysszurich.ch
Robert John, Software Engineer MYNERVA Email: robert.john@wysszurich.ch
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University, Sweden, Email: valleg@chalmers.se

Examining the effect of posture upon intracortical microstimulation evoked sensations and motor decoding (collaboration with The University of Chicago)

Brain computer interfaces assume that electrodes always evoke the same sensation when stimulated and motor neurons always act the same way for a given task. Participants, however, regularly report that the position of their arm influences the strategy they use for tasks and what they feel when stimulated. This project aims to systematically probe the effects of posture during motor tasks and the location and quality of projected fields during intracortical microstimulation (ICMS).

Description
BCIs offer an unprecedented window into understanding how the brain processes both sensory inputs and plans motor movements. We have implanted 2 arrays in somatosensory cortex (Brodmann’s Area 1) and 2 arrays in motor cortex in 2 participants at the University of Chicago. ICMS allows direct activation of cortical neurons and participants report percepts restricted to specific patches of skin and with unique qualities (buzzing, tapping, etc.) Recording from motor cortex allows us to decipher movement intentions and control prosthetic arms in the real world or in virtual reality. Our participants often insist on positioning their hands and arms in specific ways during both sensory in motor tasks and report qualitative differences in the ability to perform when their hand isn’t positioned correctly. This project aims to test the validity of their claims and find insight into how proprioceptive feedback might alter both ICMS-evoked sensations and motor plans.

Goal
The student will be guided in understanding sensory feedback with ICMS, its effects and meaning in terms of artificial perception. Additionally, the student will be taught how to decode movement intents from neural data using linear decoders. Throughout the project, the student will learn the current state of the art of neuroprosthesis with scientific literature readings.
The major goals (mandatory) for the student will be:
1. Learn the fundamentals of how bi-directional neuroprothesis function.
2. Measure the effects of posture on ICMS evoked sensations.
3. Estimate how posture alters motor intent with linear decoders.
Recommendable skills: Signal processing, neuroscience, data analysis.

Time effort required: Master project full time.
Contact Details: Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University, Sweden Email: valleg@uchicago.edu
Dr. Charles Greenspon, Senior Research Associate, Bensmaia Laboratory, University of Chicago, USA Email: charles.greenspon@gmail.com

Examining palmar and dorsal representations in human somatosensory cortex (collaboration with University of Chicago)

Intracortical microstimulation (ICMS) can be used to evoke tactile sensations and provide artificial feedback for bionic limbs. To date, most studies have focused on the sensations evoked on the palmar surface of the skin as this is most useful for grasp. Many electrodes, however, also evoke sensations on the dorsum of the hand. This is broadly inconsistent with primate literature which shown discrete palmar and dorsum regions. This project seeks to understand this inconsistency and understand how the hand is represented in humans.
 

Description
Like most sensory cortices, human somatosensory cortex is topographic – that is to say that each part of cortex represents a different patch of skin and that adjacent patches of skin are typically represented by adjacent parts of cortex. This topography holds true at large scales, but between the palmar and dorsum (front and back) of the hand, literature is inconsistent between species. Furthermore, when we stimulate human cortex with ICMS or map the hand and record neural responses, the palmar and dorsum patches are often co-located. To date, all human work has been performed on flattened 2D maps of the hand which do not capture the 3D nature of the hand and have largely ignored the dorsum. Understanding how both surfaces are represented in parallel will give insight both into how the body maps are represented across primates and indicate how ICMS should be delivered as to provide sensations as precisely as possible.
 

Goal
The student will be guided in understanding the principles of sensory feedback, its effects and meaning in terms of artificial perception, the current state of the art of neuroprosthesis with scientific literature readings, and our developed sensory-feedback system. The student will study in detail the mechanisms and principles of direct electrical nerve stimulation for sensory feedback applications.
The major goals (mandatory) for the student will be:
1. Collecting survey data where participants report the location of projected fields on a 3D app.
2. Analyzing survey data.
Recommendable skills: Signal processing, MATLAB, python, data analysis.
Time effort required: Master project full time.
Contact Details: Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University, Sweden
Email: valleg@uchicago.edu
Dr. Charles Greenspon, Senior Research Associate, Bensmaia Laboratory, University of Chicago, USA Email: charles.greenspon@gmail.com
 

Identifying the principles that provide stable and reliable control of a one-dimensional actuator in BCI - (collaboration with University of Chicago)

Identifying the principles that provide stable and reliable control of a one-dimensional actuator in BCI.
The research in motor brain-computer interfaces has been focused on providing a participant with the ability to affect objects around them. It can take the form of controlling a computer cursor, or a robotic arm that manipulates objects around them. However, the metrics of that performance have been based on overall functional success, often ignoring the precision of the movement and the ability to reproduce a specific trajectory, which are characteristics of a dexterous movement. Furthermore, they do not provide stable control that would allow for precise and small adjustments needed during the manipulation of an object, or to precisely assess the participant's intent.
In this project, we will focus on identifying the basic principles, architecture, and calibration of decoders that provide one-dimensional control using position, velocity, and acceleration, together or separate. The decoder's performance will be quantified in tasks involving precise and fast target matching, following a trajectory, with and without intracortical microstimulation.
We anticipate the project to take six months to complete.
 

Contact Details
Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University, Sweden
Email: valleg@chalmers.se
Dr. Anton Sobinov, Senior Research Associate, Bensmaia Laboratory, University of Chicago, USA
Email: ansobinov@uchicago.edu

Unveil the neural mechanisms behind the adaptation to neurostimulation - (collaboration with University of Zurich - INI)

In this project we want to unveil mechanisms behind the neural adaptation that occurs during intraneural neuromodulation and to emulate the neural process to prevent, in the real-time system.

Description
Direct nerve stimulation can be used to provide sensory feedback to upper and lower-limb amputees. The sensory feedback allowed them to discriminate, during manipulation through a prosthetic hand (bionic hand), objects of different compliances and shapes or to improve their mobility/agility while walking (bionic leg). The intensity of the elicited sensations can be modulated using the amplitude (Linear Amplitude Modulation - LAM) or frequency (Linear Frequency Modulation - LFM) of the injected stimuli. One of the drawbacks of the direct nerve stimulation is the adaptation which caused the patients to stop perceiving the restored sensations (even to low-frequency stimulation). In particular, when long-lasting trains of stimulation were delivered, the LFM stimulation (> 500 Hz) generated a fast adaptation (~ s) and can last up to tens of few minutes. In this project we want to unveil mechanisms behind the neural adaptation that occurs during intraneural neuromodulation and to emulate the neural process to prevent, in the real-time system. Although the adaptation can occur at different levels of the afferent pathway, the project goal is to investigate the neural response to Temporal Pattern stimulation at the network level. The network should simulate/emulate the adaptation times of the experimental data. Initially the network will be simulated in Brian, and afterwards it will be ported into a neuromorphic chip.
 

Goal
In this project, we want to unveil mechanisms behind the neural adaptation that occurs during intraneural neuromodulation and to emulate the neural process to prevent, in the real-time system, the fast adaptation. The adaptation can occur at different levels of the afferent pathway, we believe that three different mechanisms concur in the final adaptation: ● Spike frequency adaptation (SFA) in the afferent nerve (in the Dorsal Horn, in the Dorsal Root Ganglion) ● Short time depression (STD), at the level of the first synapses (at the interface between PNS and CNS) ● E-I network balance at level of sensory-motor cortex The three mechanisms present different time constant, SFA and STD are in the order of ms, the E-I network can reach a time constant for the adaptation of the same order as the one observed in the experiments. Although the leading mechanism is clearly the one due to the E-I network, we believe it’s worth to exploit all the free mechanisms. Once the network works on simulations, and once the new chip is available, the network will be port into the chip. The new chip, named Dynapse2, is the second generation of the Dynamic Neuromorphic Asynchronous Processors (DYNAP-SE). The new chip enhances the processing and energy-efficiency capabilities of the previous generation with the addition of a bio-signal analog front-end (AFE). It integrates analog circuits that emulate the behavior of biological neurons and synapses, and digital logic circuits for configuration.
Recommendable skills: knowledge of C++, C, embedded linux system (raspberry pi).
Extra skills: Qt programming, multithread applications.
Time effort required: Master project full time
Contact Details: Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University, Sweden
Email: valleg@chalmers.se
Dr. Elisa Donati, Lecturer at the Department of Information Technology and Electrical Engineering, Institute of Neuroinformatics, University of Zurich, Zürich Switzerland Email: eldonati@ethz.ch

Towards a prosthetic face: design a human-machine interfacing strategy for restoring sensorimotor functions in paralyzed face - (collaboration with the Maxillofacial Surgery Operational Unit of the University Hospital of Siena)

Facial movements and sensibility play a fundamental role in basic human functions such as communication, nutrition, speech and others providing a variety of cues. Neural stimulation and recording can be used to restore facial functions and improve quality of life of people with facial paralysis and/or loss of face sensibility.

​

Keywords: neural interface, neurostimulation, prosthetic face, bionics, neuroprosthetics, neural recording

​

Description

Facial palsy can be the result of central or peripheral lesions of the VII cranial nerve, which represent the major motor supply of the face. Conversely face sensibility depends on the V cranial nerve which carries tactile, thermal and pain stimuli. The lost ability to animate the face and to perceive facial l sensations has devastating complications and impaired quality of life.[1]. In addition to these psychosocial penalties, facial paralysis significantly impairs essential facial functions, including blinking, protection of the cornea, lip competence and speech, smiling, and breathing. Neurostimulation can be user to target facial muscles such as the orbicularis oculi muscle that control eye blinking[2], [3] as well as the facial nerve that can maintain overall facial muscle tone[4], [5]. On the other hand, tactile sensors in combination with neural and brain stimulation can help to restore sensibility. The development of a wearable system to restore facial muscle tone and control eye blinks would take vital steps towards being able to implement and test the effectiveness of this therapy in patients. The unique peculiarity of facial innervation which is based on a pair of nerves both for motor and sensory functions can be used to detect signals from the healthy side.

​

Goals:

The student will be guided in understanding the concepts of neurostimulation, neural recording as well as the effects of facial paralysis. They will also be introduced to state of the art of neuroprostheses with scientific literature readings, and our developed sensory-feedback system.

The major goals (mandatory) for the student will be:

1. Design a target neural interfacing strategy for restoring sensory functions in facial paralysis

2. Design a target neural interfacing strategy for restoring motor functions in facial paralysis

3. Develop an electrode placement system for facial nerves and muscle reanimation to maintain facial muscle tone + software

4. Develop an electrode placement system for restoring eye blinks, lips movements and sensations.

5. Designing an intelligent system for a closed loop system

6. Test on volunteers to obtain a statistical meaningful description of cause-effect

7. Implement the knowledge of brain stimulation

8. Develop an electrode system to detect stimuli from the healthy side and from the brain cortex and/or nerves 

Recommended skills: knowledge of C++, C, CAD, Peripheral nerve neurophysiology.

Extra skills: Qt programming, multithread applications.

Time effort required: Master project full time

​

Contact Details

Dr. Giacomo Valle, Assistant Professor, Head of Neural Bionics laboratory, Chalmers University of Technology, Sweden Email: valleg@chalmers.se  

Dr. Guido Gabriele, MD, Professor at Università degli Studi di Siena, maxillofacial and lymphatic surgeon, Email: guido.gabriele@unisi.it