Nishi Kanta Paul

Software Engineer-I @ Brain Station 23 PLC
AI, Computer Vision, Large Language Model, Medical Imaging

Dhaka, Bangladesh

Github

Google Scholar

Greetings!

I’m Nishi Kanta Paul, a Computer Science graduate from Ahsanullah University of Science and Technology (Top 7 in class), passionate about AI, computer vision, and large language models (LLMs) for healthcare and intelligent systems.

I work as a Software Engineer-I at Brain Station 23 PLC and as an Adjunct Lecturer at International Standard University, teaching Robotics, IoT, and Algorithms.

I am the Founder & Research Lead of NIMISHES Lab, an independent research lab advancing AI-driven medical imaging and intelligent healthcare systems, and a Remote Research Assistant at Washington State University, working on anomaly-aware forensic authentication under Prof. Asma Jodeiri Akbarfam. My research spans EEG sleep staging, wearable signal modelling for Parkinson’s disease, diabetic retinopathy grading, polyp and skin lesion segmentation, chest radiograph classification, and explainable forensic AI. I am lead or co-author on 9 manuscripts: 3 accepted at IEEE BECITHCON 2026 and iCONEECT 2026, 5 under review — including IEEE ICCIT 2026, IEEE JBHI, IEEE Access, and PLOS ONE — and 1 manuscript in preparation. My work on driver distraction detection has been published internationally.

With 3000+ solved problems on Codeforces, LeetCode, and CodeChef, I bring strong algorithmic and problem-solving skills. I focus on medical AI, clinical NLP, intelligent chatbots, and LLM-driven systems.

I’m always open to research or software collaborations—feel free to reach out!

Contact: nishikantapaul108@gmail.com

Recent News ⚡

September 2026	Promoted to Research Assistant at Elite Research Lab; joint first author on LCM-UNet, a 34K-parameter Mamba-based skin lesion segmentation network.
June 2026	Joined Washington State University as a Remote Research Assistant on Anomaly-Aware ForensiBlock, supervised by Prof. Asma Jodeiri Akbarfam.
April 2026	Founded NIMISHES Lab — an independent research lab advancing AI-driven medical imaging and intelligent healthcare systems.
March 2026	Joined Elite Research Lab as a Junior Research Assistant, first author on BGD-SF PolySegNet (boundary-guided polyp segmentation).
January 2026	Started as Adjunct Lecturer at International Standard University.
December 2025	Joined Elite Research Lab as a Research Intern for the MedVQA-X (Medical Multimodal AI) project.
June 28, 2025	Completed NeuroSight: AI-Powered Brain Tumor Detection.
May 18, 2025	Joined Brain Station 23 PLC as a Software Engineer-I.

Selected Publications 📄

Published Video-based Driver Distraction Detection using Transfer Learning: Bangladesh Perspective

IEEE iCACCESS, 2024 · IEEE Xplore

Transfer-learning driver distraction recognition evaluated on Bangladeshi road conditions.

Accepted SleepEffFormer: Efficient CNN-Transformer with Transition-Aware Smoothing for Single-Channel EEG Sleep Stage Classification

IEEE BECITHCON, 2026 · first author

83.9% accuracy and 78.9% macro-F1 on Sleep-EDF Expanded with 3–5× fewer parameters than comparable models.

Accepted DR-LiteNet: A Lightweight Explainable Hybrid CNN Framework for Imbalanced Diabetic Retinopathy Grading

iCONEECT, 2026 · first author

QWK 0.884 on APTOS 2019 with 4.69M parameters, outperforming ResNet50 at a fifth of the size.

Under Review AWEF-Net: An Attention-Weighted Ensemble Fusion Framework for Multi-Class Pulmonary Disease Classification

IEEE ICCIT, 2026 · first author

96.3% macro-F1 and 97.1% COVID-19 recall across 21,165 chest radiographs.

View all 9 publications →