RESEARCH & ACHIEVEMENTS
Research Focus
My research interests lie at the intersection of machine learning, image processing, and telecommunications. I'm particularly interested in developing innovative solutions that can have real-world impact in healthcare and accessibility.
Currently, I'm working on research papers focused on machine learning applications in telecommunications and sign language translation, building upon my previous work in medical image analysis.
"Research is formalized curiosity. It is poking and prying with a purpose."
Publications
Residual Block-Driven CNN for Accurate White Blood Cell Image Analysis and Classification
Presented at ICCIT 2024 Conference, Cox's Bazar
Built a Residual Block-Driven CNN machine learning model that can accurately analyze each white blood cell, classify and calculate the total number. The model achieved high accuracy in distinguishing between different types of white blood cells, which is crucial for diagnosing various blood disorders and diseases.
Abstract
White blood cell (WBC) classification plays a vital role in diagnosing various blood disorders and diseases. Traditional manual counting and classification methods are time-consuming and prone to human error. This paper presents a novel approach using a Residual Block-Driven Convolutional Neural Network (CNN) for automated WBC image analysis and classification. Our model leverages residual connections to mitigate the vanishing gradient problem and enable deeper network training, resulting in improved feature extraction and classification accuracy.
Current Research Projects
Working on journal publications
Machine Learning in Telecommunications
Exploring applications of machine learning algorithms to optimize telecommunication networks and improve signal processing. This research aims to enhance network efficiency and reliability through intelligent data analysis and predictive modeling.
Sign Language Translation Systems
Building upon my Bangla Sign Language Translator project, this research focuses on improving the accuracy and real-time capabilities of sign language translation systems. The goal is to create more accessible communication tools for the deaf and hard-of-hearing community.
Awards & Recognition
Champion in Programming Contest
Department of ETE, RUET
Became the champion in an inter-department programming competition held by the Department of Electronics & Telecommunication Engineering.
Mars Rover Project
Communication Team Leader
Currently leading the communication team for the Mars Rover project. Participated in the National Rover Challenge (NRC) with the design.
My Research Approach
Problem Identification
I start by identifying real-world problems that can be addressed through technology and research. This involves literature review and understanding the current state of the art.
Methodology Development
I design and implement innovative methodologies, often leveraging machine learning and data analysis techniques to address the identified problems.
Validation & Publication
I rigorously validate my approaches through experiments and analysis, then share the findings through conferences, journals, and open-source implementations.