Looking for a job

Md Shakib Hossain

@mdshakibhossain

AI researcher at Dhaka International University building explainable deep-learning systems for medical imaging and crop disease diagnosis.

Bangladesh

Message

At Dhaka International University, I’ve worked on deep learning for medical image analysis, including brain tumor classification across multimodal MRI data. I incorporated Grad-CAM to interpret model decisions, and my brain tumor segmentation work was accepted at ICECCT 2026.

For plant disease detection, I built a framework combining MobileNetV3-Small classification with YOLOv11n localization for chili leaf disease in smartphone-captured field images. I also developed explainable frameworks for bean and rice disease diagnosis that address distribution shifts and low-quality inputs.

My research has included collecting and curating chili and bean leaf disease datasets from real field conditions in Bangladesh. I’m preparing these datasets for public release.

I applied SMOTE-based oversampling with optimized ensemble models for industrial failure prediction, and published the findings at the International Conference on Big Data, IoT and Machine Learning. My research interests include deep learning, computer vision, and explainable AI.

Experience

Work history, roles, and key accomplishments

Deep Learning for Medical Image Analysis

Dhaka International University

Jan 2025 - Dec 2026 (1 year 11 months)

Designed and evaluated deep CNN architectures for brain tumor classification across multimodal MRI data, incorporating Explainable AI (Grad-CAM) to interpret model decisions. Developed a segmentation pipeline for brain tumor area quantification using multimodal imaging; results accepted at ICECCT 2026.

Deep Learning CNN Medical Imaging Analytics MRI Explainable AI Grad CAM Segmentation ResNet VGG EfficientNet Vision Transformer

Plant Disease Detection under Real-Field Conditions

Dhaka International University

Jan 2025 - Dec 2026 (1 year 11 months)

Built a multi-model framework combining MobileNetV3-Small (classification) and YOLOv11n (localization) for chili leaf disease detection using smartphone-captured Bangladeshi field images. Developed quality-aware and failure-aware explainable frameworks for bean and rice crop leaf disease diagnosis, addressing real-world distribution shifts and low-quality inputs.

Deep Learning Computer Vision MobileNetV3 YOLO Image Classification Object Detection Explainable AI Dataset Curation

Predictive Maintenance with Ensemble Learning

Dhaka International University

Jan 2024 - Dec 2025 (1 year 11 months)

Applied SMOTE-based oversampling combined with optimized ensemble models (Random Forest, XGBoost, LightGBM) to address class imbalance in industrial failure prediction datasets. Published findings at the International Conference on Big Data, IoT and Machine Learning (September 2025).

Machine Learning Ensemble Learning Random Forest XGBoost LightGBM SMOTE Imbalanced Learning Predictive Maintenance