My Research Work

Explainable ML for Flood Prediction in Bangladesh

Conference: International Conference on Computer and Information Technology (ICCIT) 2025. | Status: accepted

Regression-based flood prediction using SHAP & LIME for explainability.

Tools: Python, Scikit-learn, SHAP, LIME

DOI: 10.1234/flood.2026.001

View Paper →

Comparing Deep Learning Models for the Recognition of Leaf Diseases in Cucurbit Crops

Conference: International Conference on Computer and Information Technology (ICCIT) 2025.| Status: accepted

Cucurbit crops, including cucumber, melon, pumpkin, and squash, are essential to the global food supply but are highly susceptible to foliar diseases that diminish yield and quality. This study assesses six deep learning architectures - Basic CNN, DenseNet-121, MobileNetV2, ResNet-50, Vision Transformer (ViT), and Swin Transformer.

Tools: BERT, Deep Learning, ML, PyTorch

DOI: 10.1234/hatespeech.2026.002

View Paper →

Federated Graph-Transformer Synergy for Early Alzheimer’s Detection: A Dual-Modal Approach

Conference: 2026 IEEE 2nd International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN) | Status: accepted

The early diagnosis of Alzheimer’s Disease (AD) is a critical challenge in neuroinformatics, requiring the integration of complementary neuroimaging modalities such as Diffusion Tensor Imaging (DTI) and functional Magnetic Resonance Imaging (fMRI). While multimodal deep learning offers high diagnostic precision, the deployment of such models is severely constrained by data privacy regulations (e.g., HIPAA, GDPR).

Tools: Python, TensorFlow, Satellite Data, GIS

DOI: 10.1234/floodrisk.2026.003

View Paper →