Embedded Files
I work in the field of computational neuroscience. My background is in Electrical Engineering. During my master's, I worked in the field of image processing and published in several journals and conferences. Here I shall put on my recent adventure about research in my field.
Graph theory has been around for centuries and has touched several fields as a tool to analyze various systems. The main reason graph theory is a powerful tool is that it is appealing in investigating a complex system with various components interacting with each other. One such highly complex system is our human brain; hence, graph theory is being used to study the human brain ubiquitously. I am working on a few projects to unfold the mysteries of the human brain using state-of-the-art graph models.
Highlights -
My Abstract has been selected in 29th Annual Meeting of the Organization for Human Brain Mapping (OHBM 2023). This year it is being held in Montreal Canada. (13th March 2023)
Presented paper titled "Does Spatial Location of The Electrodes in EEG Matter for Tracking the Brain States?" in Twenty-Ninth National Conference on Communications 2023 at IIT Guwahati. (24th Feb 2023)
Attended Google Research Week 2023 at Ashoka Hotel, Bangalore. (29th Jan 2023)
Attended Brain Computing Learning 2023 workshop at IISc Bnaglore (7th Jan 2023)
DOES SPATIAL LOCATION OF THE ELECTRODES IN EEG MATTER FOR TRACKING THE BRAIN STATES?
In a recent paper, published in “National Communication Conference – 2023, IIT Guwahati,“ I tried to investigate a well-settled culture of using electrode position as an important factor in the EEG paradigm of scanning human brain activity. Placing electrodes is tedious, time-consuming and at times, painful to subject we are scanning using EEG. When I was to start one of my PhD project, I was pondering, "does placing electrodes precisely at every assigned location as per the protocol is “really” necessary?" After discussion with professors, I was given a green signal to investigate it. I used very basic analytical methods which are well established in literature to investigate it. The basic idea is that for ‘a task’ if we can decode the EEG signal using a subset of electrodes regardless of their location, there is hope that we do not need precise electrode positions to decode ‘that task’. Primary empirical results showed a contrast with existing beliefs around this question. We published the primary results in NCC-2023 and continue to investigate further for other important tasks like motor imagery (MI).
Demonstration of Eye Tracker at IBRO Spring School of Computational Neuroscience and AI
Title: Tracking where do we look!
I demonstrated an eye tracker and took a tutorial on why it is important to track eyes in computational neuroscience in a workshop organized in IIT Delhi. The workshop was all women workshop. It was held to motivate women in India to pursue the research work in neuroscience and science in general. I was also integral part of the organizing team.
Attended the Google Research Week 2023 at Bangalore
I attended google research week, where I learned a ton about current research culture and state of the art of computer vision, and foundational ML. I get to network with several leaders in the field and enthusiastic researchers from all over the India.
Attended Brain Computing and Learning workshop at IISc
I attended the BCL 2023 at IISc in January 2023. The theme of the workshop was Brain Computer Interface. The legendary professor Rajesh Rao gave several lectures and was part of several panel discussions. All the lectures were amazing and inspiring.
My Recent Work in IEEE Sensor Letters:
Title: Detecting Microstate Transition in Human Brain via Eigenspace of Spatiotemporal Graph
Abstract: A novel approach for analysing the EEG signal of human brain have been proposed. We have considered the EEG electrodes as nodes of the graph and correlation between the electrodes' signal as the edges. Then, using the spectral analysis of the graph, a novel method have been proposed for detecting transition in microstates of the brain. The proposed method is comprised of two steps. First, a spatiotemporal graph is constructed. Then, using eigenspace of the graph, the transition of the EEG microstate has been detected. Experimental results on publicly available dataset shows that the proposed method detect the brain state transition more accurately than state of the art.
Idea and a few Results:
My Recent Research Papers:
[1] R. Dev, S. Kumar, and T. K. Gandhi, "Detecting Microstate Transition in Human Brain via Eigenspace of Spatiotemporal Graph," IEEE Sensor Letters, April 2023. (Accepted)
[2] R. Dev, S. Kumar, and T. K. Gandhi, "Decodability of Eigenspace of EEG Graph for Motor Imagery Task," in 29th Annual Meeting of the Organization for Human Brain Mapping (OHBM 2023), 22-26 July, 2023. (Accepted)
[3] R. Dev, S. Kumar, and T. K. Gandhi, "Does Spatial Location of The Electrodes in EEG Matter for Tracking the Brain States?," Twenty-Ninth National Conference on Communications, Guwahati, 23-26 February 2023. (Accepted)
My Resume:
RESUME.pdfPage updated
Google Sites
Report abuse