Research & Projects
Identification of Biomarkers for Early Stage Hepatocellular Carcinoma (HCC)
Early stage hepatocellular carcinoma often develops silently, making timely diagnosis one of the greatest challenges in liver cancer care. This research explored how changes in gene expression can reveal molecular signatures of disease before clinical symptoms become apparent. Through bioinformatics and pathway enrichment analysis, several candidate biomarkers were identified, offering insights into how computational biology can support earlier cancer detection.
Design and Validation of a qPCR Assay for MRSA Detection
Rapid identification of Methicillin Resistant Staphylococcus aureus (MRSA) is essential for infection control and timely treatment. This research explored the development of a molecular diagnostic assay capable of detecting MRSA with greater speed than conventional laboratory methods. The work highlights how molecular diagnostics continue to transform infectious disease detection within modern healthcare.
AI Engagement for Preventive Predictive Analytics in Digital Health
Digital health platforms generate enormous potential for preventive healthcare, but only when users remain engaged long enough to produce meaningful longitudinal data. This research investigates whether adaptive artificial intelligence can improve sustained engagement through personalised insights, behavioural nudging, and conversational support. By strengthening data continuity, the study explores how engagement itself may become the foundation for more accurate preventive predictive analytics and earlier identification of chronic disease risk.
NHANES Based Early Risk Prediction for Chronic Diseases
Many chronic diseases begin years before symptoms appear, leaving valuable opportunities for prevention untapped. Using the National Health and Nutrition Examination Survey (NHANES) dataset, this research explores how demographic, clinical, laboratory, and lifestyle variables can be combined to identify individuals at elevated risk before disease develops. The work demonstrates the growing role of population scale data in enabling earlier and more personalised preventive healthcare.
Predictive Maintenance of Medical Equipment Using Long Short Term Memory (LSTM)
Medical equipment failures can interrupt patient care, increase operational costs, and reduce healthcare efficiency. This research investigates how Long Short Term Memory (LSTM) neural networks can learn patterns from historical maintenance records to predict equipment failures before they occur. The study explores the shift from reactive maintenance towards intelligent, data driven maintenance strategies within healthcare systems.
Healthcare Data Visualisation for Clinical Decision Support
Healthcare organisations collect vast amounts of data every day, yet much of its value remains hidden without effective visualisation. This research explores how interactive dashboards and analytical visualisations can transform complex clinical and operational datasets into clear, actionable insights. By making data easier to interpret, the work supports faster decision making, improved healthcare management, and more evidence based practice.