Debjoty Roy Pranto is a Research Assistant at the AVIS Research Lab and an MSc student in the Department of Robotics and Mechatronics Engineering, University of Dhaka.
He completed his BSc in the same department, where his undergraduate research focused on brain-computer interfacing and EEG signal classification.
His current research focuses on developing an AI-driven EEG digital twin framework to improve the accuracy and responsiveness of adaptive brain-computer interface systems for robotic control. The approach involves creating a virtual, continuously updated representation of a user’s neural signal patterns, enabling the system to anticipate and adapt to changes in EEG signals rather than relying solely on static, pretrained models.
The research aims to address the gap between offline EEG classification accuracy and the real-time responsiveness required for practical robotic control, with potential applications in assistive robotics and neurorehabilitation.
Debjoty co-authored the IEEE ICECTE 2026 paper “FA-EEGNet: Enhancing EEGNet with Frequency-Adaptive Kernels for the Classification of Stroke Rehabilitation Movements.” The work introduced a frequency-adaptive temporal kernel to capture inter- and intra-subject variability in EEG signals for motor imagery classification.
The study reported improvements in within-subject and cross-subject classification performance compared with standard EEGNet on the Clinical BCI Challenge-WCCI 2020 dataset.
Research Interests
Signal Processing, Brain-Computer Interfacing, Deep Learning, and Digital Twins.
Selected Publication
FA-EEGNet: Enhancing EEGNet with Frequency-Adaptive Kernels for the Classification of Stroke Rehabilitation Movements.
IEEE ICECTE, 2026.
Search for Debjoty Roy Pranto's papers on the Publications page