Khan Md Hasib

Khan Md Hasib

PhD Student @UNSW Canberra | Machine Learning, Knowledge Graph, Health Informatics, Healthcare AI

Research

My research explores the development and application of artificial intelligence and machine learning methods across healthcare, natural language processing, and computer vision domains. I am particularly interested in knowledge graph blending with machine learning and large language models to improve health systems and outcomes. My work spans medical image analysis, explainable AI, sentiment analysis, imbalanced data classification, Bengali language processing, and agricultural AI. I have published 60+ papers in top-tier journals and conferences including IEEE Transactions on Computational Social Systems, Engineering Applications of Artificial Intelligence, Scientific Reports, IEEE Access, IJCNN, and EMNLP.

1. Medical Image Analysis & Healthcare AI

Medical imaging plays a critical role in early diagnosis and treatment planning. My research focuses on building effective yet lightweight deep learning architectures for classifying and detecting diseases from medical images, including breast cancer histopathology, brain tumors from MRI, skin lesions, Alzheimer’s disease staging, and pneumonia from chest X-rays. I emphasize interpretability and clinical trust through the integration of Explainable AI (XAI) techniques.

2. Natural Language Processing & Sentiment Analysis

Sentiment analysis and text classification are powerful tools for extracting opinions and insights from textual data. My research develops machine learning, deep learning, and transformer-based models for sentiment classification, topic modeling, multi-emotion detection, and text summarization. A major focus has been on airline review analysis, where I have contributed a systematic literature review and multiple novel frameworks.

3. Imbalanced Data & Class Imbalance Solutions

Class imbalance is a pervasive challenge in real-world machine learning applications where minority classes are often the most critical. My research addresses this through novel hybrid sampling methods, cost-sensitive learning frameworks, and deep learning-based approaches for imbalanced classification. I have proposed methods such as HUSBoost and HSDLM, and published comprehensive surveys covering the landscape of imbalance handling techniques.

4. Explainable AI (XAI)

As AI systems increasingly influence critical decisions in healthcare, education, and beyond, the need for transparency and interpretability becomes essential. My research integrates explainability techniques such as SHAP, LIME, Grad-CAM, and other XAI methods into deep learning models to provide human-understandable reasoning behind predictions, particularly in medical imaging, education, and emotion recognition applications.

5. Bangla/Bengali Language Processing

Bengali, spoken by over 230 million people, remains a low-resource language in NLP research. My work addresses this gap by developing models for Bengali text classification, named entity recognition, fake news detection, speech recognition, depressive text detection, and music genre classification. I leverage pre-trained language models, transformers, and multimodal fusion techniques to tackle the unique morphological and syntactic challenges of the Bengali language.

6. Mental Health & Neurodevelopmental Disorders

Leveraging AI for early detection and assessment of mental health conditions and neurodevelopmental disorders is a growing area of my research. I develop computational methods for depression detection from social media, early autism detection through eye-tracking patterns, dementia prediction, and Parkinson’s disease prognosis using machine learning and deep learning approaches.

7. Agriculture & Plant Disease Detection

Applying computer vision and deep learning to agriculture enables automated and early detection of plant diseases, supporting food security and sustainable farming. My research develops real-time detection systems using YOLO architectures, GAN-based data augmentation with instance segmentation, and hybrid deep learning frameworks for classifying diseases in crops such as potatoes and watermelons.

8. Predictive Analytics & Forecasting

Machine learning enables powerful predictive modeling across diverse real-world domains. My research applies ML and deep learning to stock market prediction, COVID-19 forecasting, water quality assessment, traffic analysis, accident-prone area detection, crash severity analysis, residential load forecasting, and news popularity prediction, delivering data-driven insights for decision-making.

9. Deep Learning & Computer Vision

My research in deep learning and computer vision focuses on developing novel neural network architectures and applying them to diverse recognition and classification tasks. This includes deepfake detection, handwritten digit recognition, 3D gesture recognition for human-robot interaction, automatic place recognition using convolutional autoencoders, and decision tree algorithm evaluation.

10. Robotics & IoT

Bridging embedded systems with machine learning for real-world physical applications, my research in robotics and IoT includes autonomous mobile robot navigation, gesture-based human-robot interaction systems, and Arduino-based service robot prototypes controlled via Android applications.

Find me on

Google Scholar | ResearchGate | Semantic Scholar | ORCID | LinkedIn | GitHub | X (Twitter)