CV
This is a description of the page. You can modify it in '_pages/cv.md'. You can also change or remove the top pdf download button.
Contact Information
| Name | Rahul Kumar Kundu |
| Professional Title | Researcher & PhD Applicant |
| rahulkumarkundu.bd@gmail.com | |
| Location | 27/6 Dohakula, Bagherpara, Jashore, Khulna 7470 |
Professional Summary
ECE graduate specializing in deep learning and machine learning for healthcare, cybersecurity, and NLP. Actively seeking funded PhD positions for Fall 2027 to develop interpretable, practically deployable AI systems.
Education
-
2020 - 2025 Rajshahi, Bangladesh
B.Sc.
Rajshahi University of Engineering and Technology (RUET)
Electrical and Computer Engineering
- Ranked 7th out of 60 students in the department.
- Last 60 Credits CGPA 3.91.
- Thesis: ED-Net — a hybrid deep learning framework for polyp segmentation integrating EfficientNetB0 with Double U-Net, ASPP, and SE-block attention.
Interests
Publications
-
2025 ED-Net: A Hybrid Deep Learning Framework for Precise Polyp Segmentation Integrating EfficientNetB0 with Double U-Net Architecture and Atrous Spatial Pyramid Pooling with Test-Time Augmentation
Biomedical Signal Processing and Control, Elsevier (Q1, IF: 4.9) — Under Review
Achieved 97.13% accuracy, 92.26% Dice, and 91.08% IoU on Kvasir-SEG with TTA, outperforming recent SOTA models including CSwinDoubleU-Net and SwinSAM.
-
2025 Beyond the Leaf: Intelligent Lightweight Disease Detection in Tea Plants with Spatial Convolutional Neural Networks and Explainable AI
2025 IEEE 7th International Conference on Sustainable Technologies For Industry 5.0 (STI), IEEE Xplore
Lightweight CNN-based tea leaf disease detection with Grad-CAM explainability. Published in IEEE Xplore.
-
2025 Advancing Cybersecurity: A Study on Phishing Website Detection Using Machine Learning
2025 8th International Conference on Computer and Information Technology (ICCIT), IEEE — Accepted
Ensemble phishing detection system (XGBoost + LightGBM + CatBoost) with SHAP-based interpretability achieving 93.71% accuracy and 99% phishing recall.
Projects
-
MetaSeg — Metadata-Integrated Skin Lesion Segmentation
A deep learning segmentation framework integrating patient demographic metadata through FiLM modulation for improved skin lesion segmentation across diverse patient populations.
- FiLM-based metadata fusion into decoder layers.
- Collaboration with Qatar University.
- Targeting Computers in Biology and Medicine (Elsevier Q1).
-
Pragmatic-Aware Emotion Grounding (PAEG) for Bangla
Investigates pragmatic feature loss in machine-translated Bangla emotion datasets. Proposes a framework combining lexicon retrieval and culturally-grounded augmentation to improve LLM emotion recognition for low-resource Bangla.
- Three self-built datasets totaling approximately 26,000 samples.
- Targeting Information Processing and Management (Elsevier Q1).
-
RUET-ASSR Official Website
Led a team to build the official website for the Astronomy and Science Society of RUET using Next.js, Prisma, and MongoDB.
Skills
Academic Leadership and Service
Awards
-
2024 Technical Scholarship — University Grant Commission (UGC)
University Grant Commission, Bangladesh
Awarded for three consecutive years (2022, 2023, 2024) to top-ranked students based on B.Sc. annual results.
-
General Grade Scholarship — SSC and JSC Examinations
Government of Bangladesh
Awarded to nationwide highest-ranked students in Grade 10 (SSC) and Grade 8 (JSC) examinations.