Funded Research

We will apply machine learning to pre-existing data to determine which factors best predict individual response to cerebellar neuromodulation. Then we will leverage these data to design a protocol for individualized, patient-centered, paired neuromodulation-therapy for cerebellar patients. The patient outcome data will continuously be fed back into the predictive models to promote both scientific and therapeutic advancement.

We will conduct a clinical trial of thalamic deep brain stimulation in a cohort of patients with cerebellar disorders guided by critical insights gained from subacute implantation of externalized stimulating and recording electrodes. To provide a more informed approach to targeting invasive neuromodulation, we propose to create a “thalamic map of cerebellar function” using multimodal mapping and behavioral characterization of patients with thalamic interventions such as deep brain stimulation and focused ultrasound as well as non-invasive neuromodulation in people with cerebellar disorders.

The RCP’s mission is to create an adaptive neuromodulation system that records brain activity and administers stimulation to improve symptoms in patients with cerebellar disorders. For this mission to be successful, we need to know where to record and where to stimulate. Identifying these locations for current neuromodulation therapies required decades of trial and error. Our proposal will identify therapeutic targets for specific symptoms of cerebellar dysfunction within 5 years.

This proposal will uncover the cerebellar physiology underlying human ataxia and will investigate the therapeutic effects of applying stimulation to the dentate nucleus and to the interposed nucleus. Ten individuals will provide informed consent. Each participant will be genetically confirmed to have spinocerebellar ataxia type 6 (SCA6) which is associated with a clinical syndrome and the pathological loss of cerebellar Purkinje cells. This project will be coordinated and executed by two expert clinical (ataxia) and research (DBS) centers of excellence. The groups will synergize and combine neurosurgical, neurological, neurophysiology, neuromodulation, and neuroimaging expertise. This project aims to fill the gaps in our current understanding of ataxia physiology and DBS. The project will deliver an adaptive DBS trial to address a human cerebellar disease.

At CHOC, we record and stimulate up to 12 temporary epilepsy-type depth electrodes with 16 channels each to identify targets for DBS in children. Electrodes are placed in multiple regions, including inputs and outputs from cerebellum, and related to movement kinematics and muscle activity. We propose to use machine learning, robotics, control theory, human electrophysiology, and human psychophysics to learn and test feedback controllers that modify cerebellar output to improve movement and cognition in children.

As the AI core lead for the RCP, we will develop a foundational system that: i) can relate neural activity to behavioral and symptoms states that are impacted in cerebellar disorders, and ii) can predict the effect of neuromodulation on the relevant neural-behavioral responses. Furthermore, we will use these models to develop feedback-driven neuromodulation systems that can regulate abnormal neural activity patterns and symptom states in cerebellar disorders. Collectively, our project Aims will enable a paradigm shift in neuromodulation for cerebellar disorders by enabling intelligent AI-directed neuromodulation for cerebellar disorders.

We are designing methods for spatiotemporal electrical microstimulation (uStim) to drive neural population activity to specified states. More specifically, we are applying uStim using a 96-electrode Utah array in non-human primates and recording spiking activity from a population of neurons, which is either on the same array and/or on a nearby array. We select a goal for the population activity state, and the key challenge is to design the stimulation parameters: which electrodes to stimulate, at what times, and with what waveform, to achieve the goal state. We will perform these experiments in prefrontal cortex. This is because we already have an extensive understanding of how these techniques work in this area, and relevantly for RCP, because insights we gain here are relevant for the cognitive aspects of cerebellar dysfunction.

• Soldado-Magraner, Joana, et al. "Robustness of working memory to prefrontal cortex microstimulation." Journal of Neuroscience (2025)

• Minai, Yuki, et al. "MiSO: Optimizing brain stimulation to create neural activity states." Advances in Neural Information Processing Systems 37 (2024): 24126-24149

Our project will restore gait and balance through a neuroprosthetic that combines advanced neural recording, stimulation technology, and AI-driven closed-loop neuromodulation. We will develop an NHP model of cerebellar dysfunction, create AI models to predict cerebellar output, and implement the prosthesis to bypass the diseased cerebellum using M1 and STN recordings that are ‘readily’ transferable to humans using state-of-the-art electrodes. We will utilize a multi-modal closed-loop neuromodulation strategy to enhance treatment for patients with cerebellar disorders.

Two critical parameters necessary for development of closed-loop neuromodulation for cerebellar dysfunction are knowing where and when to stimulate. A prime candidate for “where”, is the thalamus, a key output hub for the cerebellum. Neuromodulation of the thalamus has already shown success across several diseases ranging from Tourette syndrome to epilepsy to pain, and at least one cerebellar disorder, essential tremor. Additionally, the thalamus has substantial connections across motor, cognitive, and emotional circuitry suggesting an ideal location for neuromodulation to improve dysfunction across these domains. However, the specific contributions of thalamic nuclei to each of these domains, and what stimulation strategy is likely to improve function, will require careful mapping.  

My team at the University of Colorado (CU) in collaboration with the group at UT Southwestern (Nader Pouratian, Peter Tsai), propose to accelerate mapping of thalamic function as it relates to cerebellar output by increasing patient numbers, recording contexts, and biomarker subspace exploration for closed-loop neuromodulation. We will improve our understanding of the role of thalamic nuclei in online control and correction of motor, cognitive, and emotional functions through electrophysiologic mapping, concurrent with stimulation and task-based perturbations. These findings can be leveraged for neuromodulation targets across both location and timing for closed-loop stimulation.  

Innovative transcranial ultrasound approaches have the potential for real-time, closed-loop correction of pathologic neurological responses. Compared to Deep Brain Stimulation (DBS), substantial advantages include higher dimensionality (highly complex assessment and stimulation patterns) as well as the potential for compact portability and low cost, without surgery. This project aims to develop a fully integrated ultrasound system for closed-loop neuromodulation and high-resolution imaging in awake animals within two years and then validate and advance the technology for translation.

Dr. Hammer is a movement disorders neurologist and neural engineer from Damascus, Maryland, with subspecialty expertise in deep brain stimulation. Her research bridges engineering, neuroscience, and clinical medicine to develop novel neuromodulation therapies for movement disorders, particularly those involving the cerebellum.

With support from the Raynor Cerebellum Project, she is leading a two-year study to develop and validate precise, quantitative measures of ataxia symptoms using wearable sensors and tablet-based tasks. By tracking symptom severity and day-to-day variability in people with hereditary ataxia, her team is building the critical tools needed to measure symptoms accurately and apply them in future deep brain stimulation trials for cerebellar disorders.

Dr. Carina Oehrn is a physician‑neuroscientist at UC Davis, where she leads a lab at the Center for Neuroscience and the Department of Neurological Surgery. Her team specializes in human intracranial recordings and deep brain stimulation, with a focus on the neural networks underlying non‑motor symptoms in movement disorders. Her goal is to harness neural signals to develop new neuromodulation therapies for cognitive and affective symptoms using frequency‑based stimulation strategies and adaptive DBS.

Dr. Oehrn’s current project examines cerebellar–cortical interactions in patients with cerebral palsy. By directly recording brain activity from implanted cerebellar electrodes combined with EEG, her research seeks to identify neural signatures of non‑motor functions in these networks and test new brain stimulation strategies, paving the way toward novel therapies for cerebellar disorders.

She earned her PhD at the University of Bonn in Germany and completed postdoctoral training in the Starr Lab at UCSF.

‍

Our overarching objective is to develop a DBS system for motor and non-motor symptoms of cerebellar impairment, building on our prior experience with electrophysiology and stimulation of the cerebello-thalamo-cortical networks for acquired brain injury. We will start the investigation with invasive targeting of the dentate cerebellar nucleus, alongside non-invasive recordings of the cortex and musculature. At this initial phase, we will iteratively advance the electrophysiological methods in order to maximize learning for technology development. Consistent with our prior investigation, this phase of research will establish a) the therapeutic benefits of DBS for motor symptoms, b) an assessment of stimulation effects relative to dentate nucleus topography, c) electrophysiological investigation of cerebello-cortical interactions, d) development of biomarkers for optimizing DBS programming, and e) investigation of electrophysiological correlates and markers of non-motor domains. 

Dr. Nettekoven is a cognitive neuroscientist whose research aims to advance non-invasive brain stimulation for cerebellar disorders by combining cutting-edge imaging and stimulation techniques. Using highly sensitive brain recordings (OPM-MEG) and precise ultrasound stimulation guided by individualized brain maps (fMRI), she seeks to better understand cerebellar function and develop targeted, closed-loop treatments to improve motor and cognitive symptoms in patients.

Dr. Nettekoven is a strong advocate for open and equitable science. She shares her lab’s tools and methods and runs a blog demystifying neuroimaging methods. She organizes symposia and educational events that highlight the roles of the cerebellum to better reflect the cerebellum’s importance in cognition, development, and healthy brain function.

Dr. Nettekoven received her BSc at Heidelberg University and completed a DPhil at the University of Oxford, where she identified a link between cerebellar neurochemicals and movement. In 2025, she returned to Oxford as a Wellcome Early Career Fellow. www.caroline-nettekoven.com