Kumar's Research Group

Overview of Current Research

My lab's interdisciplinary research lies at the intersection of neuroscience, control theory, and machine learning, aiming to develop novel modeling and closed-loop adaptive control approaches that accelerate next-generation electrical stimulation-based therapies for neurological, psychiatric, and cardiovascular diseases. Currently, most electrical stimulation technologies are open-loop, and various stimulation parameters are manually titrated by expert clinicians, resulting in the treatment’s non-optimal efficacy and severe side effects post-therapy, including multiple visits to clinicians to adjust stimulation parameters. Closed-loop (or adaptive) electrical stimulation, utilizing optimal feedback control principles, holds great promise in achieving the desired therapeutic efficacy while minimizing side effects, thereby paving the way for the development of next-generation intelligent electrical stimulation-based therapies for brain and cardiovascular diseases. This is a cutting-edge topic with high interest in neurotechnology.

Current Research Projects

Closed-Loop Deep Brain Stimulation for Controlling Neural Oscillations and Synchronization

Neurophysiological and psychiatric disorders lead to severe cognitive impairments in patients and have been attributed to pathological neural synchronization and abnormal oscillations in brain circuits associated with specific diseases. Our lab focuses on developing a closed-loop DBS computational framework for rigorous testing of DBS strategies and designing real-time, feedback-based, multi-input, multi-output (MIMO) DBS techniques to provide a safe and long-lasting effect on the desynchronization of pathological neuronal activity in Parkinson’s disease. We use dynamical systems theory, control theory, machine learning, and computational neuroscience to achieve our goals.

Recent Publications:

Aanuoluwapo V. Olumuyiwa and Gautam Kumar. Proportional-Integral Controller-Based Deep Brain Stimulation Strategy for Controlling Excitatory-Inhibitory Network Synchronization. In Proceedings of the 2025 American Control Conference. Denver, Colorado, July 8-10, 2025, pp. 4627-4634
Joseph Schmalz, Rachel V. Quinarez, Mayuresh V. Kothare and Gautam Kumar. Controlling Neocortical Epileptic Seizures using Forced Temporal Spike-Time Stimulation: An In Silico Computational Study. Frontiers in Computational Neuroscience, 2023:2023-06
Joseph Schmalz and Gautam Kumar. Controlling synchronization of spiking neuronal networks by harnessing synaptic plasticity. Frontiers In Computational Neuroscience, 13(61):1--17, September 2019

Prediction of Human Gait Events Using Deep Learning Models

Gait impairment is a debilitating symptom of Parkinson’s disease (PD), profoundly affecting patients’ quality of life. While deep brain stimulation (DBS) has been highly effective in treating PD symptoms, such as bradykinesia, rigidity, and tremor, its impact on gait in PD patients is controversial, with some studies reporting improved gait while others report worsening. In collaboration with Professor Doris Wang, Ph.D. from UCSF, our research team is developing deep neural network models to predict the onset of gait events in PD patients using neural data with DBS. Developing computational models that predict the onset of gait events under various DBS parameter settings would be a promising direction for optimizing DBS parameters to best improve patient-specific gait. Additionally, developing such models would facilitate adaptive closed-loop DBS for gait improvement.

Human Emotion Using Electroencephalography (EEG)

Our lab is interested in developing predictive models for classifying and predicting human emotions using electroencephalography (EEG) data. We are currently developing our electroencephalography (EEG) lab to study human emotions and collect brain data.

Closed-Loop Auditory Stimulation

Our lab is also interested in developing closed-loop predictive control-based auditory stimulation technologies to alter human emotional states and stress. We are particularly interested in investigating the effect of Binaural Beats (BB) on emotion and stress, and whether BB can alter induced emotional states using auditory stimuli and reduce induced stress during MIST-style arithmetic tasks.

Interested in Joining Our Lab

We are actively looking for young researchers (undergraduate and graduate students) to join our lab and experience interdisciplinary research. If you are interested, please send a brief email with your interests in our research to Dr. Kumar at gautam.kumar@sjsu.edu.

Funding

NIH R16: Closed-Loop Deep Brain Stimulation for Controlling Abnormal Neural Synchronization and Oscillations. (July 02, 2024 – June 30, 2028); PI: Gautam Kumar; Amount: $586,000.
College of Engineering Faculty Small Group Project: Control Theoretic Framework for Design and Analysis of Closed-Loop Brain Stimulation. (January 01, 2022 - December 31, 2023); co-I: Gautam Kumar; Amount: $49,316.
NIH SPARC: CONTROL-CORE: A Modular Simulation Environment for Design and Prototyping of Closed-loop Peripheral Neuromodulation Control Systems using the O2S2PARC Platform. (September 15, 2020 - September 14, 2023); co-I: Gautam Kumar; Amount: $171,984.
Murdock Partners in Science (M. J. Murdock Charitable Trust): Deep Learning-based Noninvasive Brain Decoder. (March 20, 2019 – May 01, 2022); PI: Gautam Kumar; Amount: $15,000.
University of Idaho Seed Grant: Harnessing Neuroplasticity for Rewiring Brain Circuits. (May 15, 2017 – August 31, 2018); PI: Gautam Kumar; Amount: $11,915.