Submitted Papers

Key Technologies : Graph Neural Network , ECG Signal Pre-Processing , Federated Learning , EEG Data Analysis 

Contributions : First Author

Outcomes : The FedGNN project likely achieved significant improvements in ECG signal identification accuracy by combining Graph Neural Networks with clinical features in a privacy-preserving, decentralized federated learning framework. This approach potentially enabled collaborative model training across multiple healthcare institutions without compromising patient data privacy, while also demonstrating enhanced scalability and efficiency in handling large-scale, multi-institutional ECG datasets. 

Journal: IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING (TBME)

Key Technologies : Image Pre-processing , Clinical Handcrafted Features , Developed a Decision Support System

Contributions : First Author

Outcomes : This research presents an intelligent decision support framework for analysing malignancy patterns in ovarian cancer using features derived from ultrasound imaging. By leveraging advanced image processing and machine learning techniques, the system aims to enhance the accuracy and efficiency of ovarian cancer diagnosis, potentially improving early detection rates and treatment outcomes. 

Journal: Neural Computing & Applications (Springer Link)

Key Technologies : Graph Neural Networks , Graph Implementation , Handcrafted Features Extraction Methods

Contributions : First Author

Outcomes : The CHXGNN initiative likely improved the accuracy of respiratory disease detection by combining Graph Neural Networks with clinical handcrafted features, effectively utilizing structural patterns in chest X-ray images. This advanced framework offered a robust and interpretable diagnostic tool for clinicians, enhancing the detection and classification of various respiratory conditions. 

Journal: Computers in Biology and Medicine (Elsevier )

Key Technologies : Explainable AI , Traditional Machine Learning , Utilising Histopathology Data

Contributions : First Author

Outcomes : This project developed an explainable AI system for breast cancer grading, enabling transparent and accurate classification of histopathology images. By providing clear explanations for its decisions, the model likely enhanced trust in AI-assisted diagnoses while maintaining high accuracy, potentially improving clinical decision-making and patient care in breast cancer management. 

Journal: Engineering Applications of Artificial Intelligence (Elsevier )