My Projects
A showcase of my work across various technologies
All3D ModelingArtificial IntelligenceBangla DialectsCC++CSSChest X-Ray Image ProcessingData AnalyticsData PreprocessingDeep LearningExplainable AIGLSLHTMLJavaJava SwingJavaScriptKerasLIMEMachine LearningMySQLNamed Entity RecognitionNatural Language ProcessingNode.jsOpenGLPHPPandasPythonSARIMASARIMAXSMOTESQLTensorFlowThree.jsTransfer LearningVGG19WebGLiGraphics
Forestscape
A dynamic forest environment simulation with seasonal transitions and interactive camera controls built using custom shaders.
Three.jsGLSLWebGLNode.js+1
- Implemented custom GLSL shaders to enhance lighting realism and surface depth
- Built fully interactive 3D forest with seasonal animations and perspective camera movement
Butterfly Optimization and Deep Learning to Classify Heart Sound Signal
An AI-based system using deep learning and Butterfly Optimization Algorithm to classify heart sound signals for early cardiovascular disease detection.
PythonArtificial IntelligenceDeep Learning
- Developed AI models for accurate heart sound classification using deep learning
- Extracted and analyzed time-frequency features from PCG signals
Transfer Learning and Explainable AI for Brain Tumor Classification
An AI-based system using transfer learning and explainable AI techniques to classify brain tumors from MRI data, improving diagnostic accuracy in resource-limited settings.
PythonDeep LearningTransfer Learning
- Collected and processed MRI brain tumor data from multiple Bangladeshi hospitals
- Implemented transfer learning models VGG16, VGG19, and ResNet50 for classification
A Comprehensive Analysis of COVID-19 Detection Using Bangladeshi Data and Explainable AI
An AI-powered system using machine learning, deep learning, and explainable AI techniques to detect COVID-19 from chest X-ray images with high accuracy.
PythonMachine LearningDeep LearningTransfer Learning+5
- Utilized a large dataset of Bangladeshi chest X-ray images with four diagnostic classes
- Implemented machine learning, deep learning, and transfer learning models for classification
Unveiling the Genetic Mosaic: A Multi Model Exploration of Rice Varietal Diversity
A pattern recognition project exploring Turkish rice grain diversity using CNNs and transfer learning with 99.9% accuracy.
PythonTensorFlowKerasTransfer Learning
- Processed 75,000+ grain images to analyze genetic diversity in Turkish rice
- Achieved up to 99.9% classification accuracy with CNNs and transfer learning models
Advanced Data Analytics and Time Series Forecasting for NYC Taxi and Dhaka Accident Dataset
Time series forecasting and geospatial analysis on NYC taxi and Dhaka accident data for urban safety and traffic insights.
PythonSARIMASARIMAXData Analytics
- Analyzed over 2 major urban datasets for NYC and Dhaka accidents and mobility trends
- Applied SARIMA/SARIMAX models for accurate time series forecasting of taxi demand
ANCHOLIK-NER: A Benchmark Dataset for Bangla Regional Named Entity Recognition
A linguistically diverse NER dataset developed for Bangla regional dialects, covering Sylhet, Chittagong, and Barishal.
Natural Language ProcessingNamed Entity RecognitionBangla Dialects
- Compiled a benchmark NER dataset for three major Bangla regional dialects
- Collected and cleaned over 10,000 sentences from various public and online sources
Dental Care Appointment System
A group project creating a dental appointment system using PHP, CSS, and JavaScript.
PHPCSSJavaScriptSQL+1
- Developed a functional appointment scheduling system with backend support
- Implemented user-friendly interfaces with responsive CSS styling
Hyacinth: Pharmacy Management System
A Java Swing and MySQL-based system to manage pharmacy operations including inventory, sales, purchases, and expiry tracking.
Java SwingMySQLJava
- Developed a complete pharmacy inventory management module
- Implemented sales and purchase tracking with database integration
Death Stalker
A 2D Metroidvania-style game developed using C/C++ and iGraphics library.
CC++iGraphicsOpenGL
- Designed and implemented 2D sprite animations for characters and enemies
- Developed collision detection system for player, enemies, and environment
Brain MRI Classification on the Context of Imbalanced Data of Bangladeshi Sample
A deep learning approach for classifying brain MRI data using CNN with SMOTE to handle class imbalance in Bangladeshi medical datasets.
PythonDeep Learning
- Developed a CNN model to classify brain MRI images from a Bangladeshi dataset
- Applied SMOTE to effectively address class imbalance in medical image data
Brain MRI Classification on the Context of Imbalanced Data of Bangladeshi Sample
CNN-based classification model addressing class imbalance in Bangladeshi brain MRI datasets using SMOTE.
Python
- Built a CNN classifier for brain MRI tumor categorization on Bangladeshi medical data
- Applied SMOTE to mitigate effects of class imbalance in minority tumor categories
Diabetes Prediction using Machine Learning
A machine learning model that predicts the likelihood of diabetes using health and lifestyle data.
PythonMachine LearningPandas
- Preprocessed health-related data and handled missing values
- Trained multiple ML models including Logistic Regression, SVM, Random Forest, and XGBoost
Heart Disease Prediction using Neural Networks
A neural network model built using Keras and TensorFlow to predict the presence of heart disease based on clinical data.
PythonTensorFlowKerasPandas+2
- Built a deep learning model using TensorFlow backend and Keras layers
- Used Cleveland Clinic heart disease dataset for training and validation
Lung Cancer Prediction using Machine Learning
A machine learning pipeline for predicting lung cancer from clinical and lifestyle data using various classification models.
PythonMachine Learning
- Collected and loaded lung cancer dataset from Google Drive into Colab
- Performed data preprocessing, feature selection, and label encoding
Heart Disease Prediction using Machine Learning
An AI-powered system for predicting heart disease from patient health indicators using various machine learning models.
PythonPandasMachine LearningData Preprocessing
- Used a comprehensive dataset containing demographic and clinical features to train ML models
- Preprocessed data by encoding categorical variables and splitting into train/test sets
Stroke Prediction using Machine Learning
A machine learning project that predicts stroke from patient records using various classification models, with performance comparison before and after handling class imbalance.
PythonPandasMachine Learning
- Explored healthcare dataset with 5,110 entries across 11 health-related features
- Performed EDA with bar plots, countplots, and correlation heatmaps to understand class and feature distributions