Historically, brain machine interfaces (BMIs) have traded high spatial resolution for limited cortical coverage. Micro electrode arrays (MEAs) can record from single neurons and capture full spike waveforms but can only record from a small, mm scale, portion of the brain. Electrocorticography (ECoG) and electroencephalography (EEG) can record from multiple cortical sites but integrate the activity of many neurons together, obscuring the intricacies of the underlying circuits. We are developing implantable devices that bridge the gap between resolution and scope. Specifically, we are designing a platform technology for accessing a set of independently addressable neural interrogators or nodes distributed across the cortical surface. The system will enable clinicians and researchers to place nodes anywhere on the cortex, agnostic to location and therefore to the specific needs of the treatment or experiment being performed. Nodes will enable simultaneous read and write capabilities and can be used to track the flow of information across the brain. Multiple stages of wireless powering and data transmission not only eliminate percutaneous elements, preventing infection and enabling full mobility for the subject, but also eliminate transdural electrical cabling. Therefore, nodes can move with the brain, untethered to the skull, and may be placed independently, alleviating many of the surgical challenges currently faced while implanting MEAs. With a single device, neuroscientists and physicians will be free to specify a custom BMI suited to their needs, not limited by resolution, small recording area, or even interface modality.
Many, if not all, tissue interface related failures of microelectrode arrays are caused by oxidative stress. The neurofilm project aims to treat oxidative stress through local release of anti-inflammatory and antioxidant drugs to the tissues surrounding neural implants. Through polymer-drug interactions, we are developing a thin film which can stabilize and extend the release of these drugs to locally modulate the foreign body responses at the cellular level, without causing adverse systemic effects. We hope that this treatment will ultimately further basic neuroscience research, as well as current neural prosthetic treatments, by increasing reliability and functional recording lifetime of microelectrode arrays.
Epidural electrical stimulation of the spinal cord is a therapy currently approved for patients diagnosed with intractable pain, but it is also able to engage the neural circuitry involved in locomotion. However, the mechanisms by which this happens are not fully understood. My research uses computational modelling to understand these mechanisms and engineer new therapies for patients with altered locomotion. First, we are using several imaging modalities to extract personalized three dimensional models of the neural structures recruited by stimulation. Second, we use the finite element method to compute the electric potential distribution created by the stimulation. Finally, we are creating biophysically realistic models of the neural circuits responsible for locomotion in the spinal cord. This platform allows us to prototype novel stimulation paradigms in sillico, before progressing to experiments in vivo and eventual human translation.
Essential tremor (ET) is the most common adult movement disorder (Louis, Ottman et al. 1998), with a greater than 20x prevalence than Parkinson’s disease. The predominant symptom of ET is typically a kinetic tremor of the hands, but symptoms also commonly include tremor of the head and voice. However, the mechanisms underlying ET are not well understood, and no curative treatment exists. Prescription medications may be effective for mild or moderate tremor, while others self-medicate with alcohol, but for patients with severe, medication-refractory tremor, deep brain stimulation (DBS) is a valuable treatment option and is usually preferable to thalamotomy. DBS devices are typically left powered on continuously throughout the day and night to control the symptoms of ET, and many patients experience side effects of stimulation, such as dysarthria, paresthesias, and gait ataxia. In the current form of clinically available DBS therapy, a physician observes patient symptoms and adjusts stimulation parameters (e.g. voltage and frequency of the electrical stimulation) at appointments that may be days, weeks, or months apart. In contrast, a system that can sense biomarkers and automatically adjust stimulation parameters could respond to changes in the patient's behavior, goals, or disease state with a response time on the order of seconds. Specific biomarkers might include neural activity related to movement initiation assayed from electrocorticography (ECoG) or electroencephalography (EEG), and wearable kinematic sensors (e.g. inertial motion units, IMUs, common to many consumer wearable electronics). Such functionality would open the door to a broad class of biomarker-controlled closed-loop DBS paradigms with the potential to more effectively treat the symptoms of ET, reduce stimulation-related side effects, reduce the power consumption of the implanted device and accordingly reduce the frequency of surgical battery replacements.
The emergence of electrical epidural spinal cord stimulation (SCS) has provided a promising avenue for the treatment of a variety of sensorimotor disorders. Unfortunately, the way in which SCS impacts neural circuitry in the spinal cord as well as upstream networks in the thalamus and cortex is poorly defined. This creates a problem as therapeutic implementation can be inefficient and lead to unpredictable or adverse results. Using both computational and in vivo experimental methods, my project focuses on how SCS in cervical spinal cord affects underlying neural circuits and their relation to the input and output of the upper limbs. Our first aim is to build a detailed, biophysically realistic computational model that describes the flow of information as it travels from the periphery to the cortex, and back. In conjunction with in vivo studies, this model will help us understand the mechanisms by which SCS acts on the spinal cord, and allow for accurately parameterized stimulation therapies. Our second aim is to use this framework to develop a brain-spinal interface. By simultaneously recording neural activity in the cortex, we can generate a closed-loop that will adjust spinal stimulation based on decoded cortical signals. This design will enable an adaptive therapeutic that will optimally adjust stimulation when necessary, decreasing the chance of unwanted side-effects.
We are designing a miniature, implantable, Bluetooth-based sensing device for the benefit of untethered, free-moving animal research. We primarily use off-the-shelf components, and push the size and power to their limits as much as possible. Here are some features of the device: 1) Data transmission: Bluetooth Low Energy (or BLE). BLE is an extremely low power and compact wireless protocol, dedicated for transmitting small, infrequent pieces of data. Estimate data rate is 20ksps, with a typical current consumption of 1mA (100Hz). 2) Sensors: We will primarily use two categories of sensors: kinematic sensors (or IMU, Inertial Motion Unit) and muscle sensors (force sensor and EMG sensor) Kinematic sensors together with data fusion algorithm will give us absolute orientation at every time moment. We can monitor individual joint movements (flexion, extension, rotation) and construct whole-body kinematics. This would be a complementary approach to the current camera tracking system. 3) Wireless charging: Estimate power delivery is 10mW. We will customize the receiving coil so that it can fit into our device. We will build a rectifier and regulator on board.
TRANSFORM DBS stands for Trans-diagnostic Restoration of Affective Networks by System Identification and Function Oriented Real-Modeling and Deep Brain Stimulation. The initiative aims to develop a closed-loop DBS implant for the treatment of intractable psychiatric illness, specifically to address the needs of returning veterans. A tiny implanted, programmable device will record neural activity from multiple areas of the brain, sense pathological activity, and deliver targeted electrical stimulation to relieve symptoms. DBS is already successful for the symptomatic treatment of Parkinson’s disease and essential tremor. TRANSFORM DBS aims to expand the DBS approach to address psychiatric illness, including post-traumatic stress disorder, traumatic brain injury, depression, anxiety, substance abuse, and pain. Nicole focuses on identifying neural signatures of psychiatric illness and how DBS in different brain areas affects associated neural circuits.
BioLuminescent OptoGenetics (BL-OG) holds the promise for inter-luminescence, the activation of microbial opsins on one cell by the bioluminescent proteins expressed on a different cell. Our project aims to establish a cell culture based inter-luminescence setup to assess the influence of distance on the effects of BL-OG. If possible, it would not only provide a proof of concept for the process across cellular distances, but also a versatile in vitro method for the study and testing of inter-luminescence. Such an approach, performed under conditions which can be matched to physiologically relevant contexts, could open the way to potential therapeutic applications in which cellular communication and interaction is key, such as spinal cord injuries and motor system disorders, two major research areas of this laboratory.
The field of retinal prosthetics faces several challenges in restoring visual acuity, including electrical field interference, microfabrication, electrode proximity, and high stimulation thresholds. This project aims to address these challenges by stimulating neurons cultured on an electrode array that synapse directly onto the retina. After reviewing retinal prosthetic initiatives currently in development and designing a model for the stimulator, we are examining whether the stimulation of cultured neurons synapsed onto retinal sections and cultures evokes responses and induce plasticity with the eventual goal of testing this system in vivo.
Question: Are our polymers compatible with, and non-toxic to, neural tissue. Because we are designing and synthesizing custom polymers for this project, there is no literature to indicate cellular viability in the presence of these polymers. In order to determine if our polymeric design will be non-toxic to neurons in vivo, this project will test the polymer on cells in vitro. You will culture cells on the surfaces of several polymer substrates and monitor cellular viability. This project can expand in several different directions, depending on the interests of student.
We are developing a genetic manipulation technique to reversibly and controllably shut down sensation in the hindlimb of rodents. Students will learn and apply cutting edge methods in viral transfection of neural tissue, electrophysiological recording and neural data analysis. The goal of this project is to screen several engineered ion channels for the ability to shut down signals from neurons in the dorsal root ganglia, the primary sensory afferents that carry information from the periphery to the central nervous system. We will create a novel pre-clinical model that mimics the situation of several patient populations, such as lower limb amputees, and allows us to engineer electrical interfaces to alleviate the deficit. Hands on experience with laboratory animals will be a major component of the research experience.
We are planning to use a pair of magnetometers to keep tracking of a free-moving magnet in 3D space. We will reconstruct its trajectory by converting the magnetic field data we measured back to position data, with the aide of optimization algorithms. You will work to minimize noise in the data through the implementation of a Kalman filter and other high-pass filters. Additionally, you will learn to solve for local and distant minima of multi-variable function with iterative approaches including Particle Swarm Optimization