Advancing Meibomian Gland Dysfunction Management: Clinical Practice Patterns, Thermomechanical Therapy and Imaging Innovations

Meibomian gland dysfunction (MGD) is a primary contributor to evaporative dry eye disease (DED) and continues to present diagnostic and therapeutic challenges in eye care. The present thesis critically evaluates contemporary approaches to MGD, investigates novel therapeutic outcomes and develops an automated tool for objective meibography analysis. The literature review synthesised evidence regarding the pathophysiology, diagnosis and management of MGD. The review identified gland obstruction, lipid deficiency and ocular surface inflammation as central factors, and described emerging technologies for imaging and treatment. A cross-sectional survey of practicing eye care practitioners assessed current clinical approaches to DED and MGD. The results indicated variability in diagnostic testing and management strategies, with substantial reliance on symptomatic assessment. Identified barriers to optimal care included restricted access to advanced imaging and inconsistent application of evidence-based therapies. A retrospective analysis of thermomechanical treatment using Tixel® (Novoxel, Netanya, Israel), for MGD was performed with clinical records from South Africa. The outcomes demonstrated statistically significant improvements in both subjective symptoms and objective ocular surface tear film measures, supporting its potential safety and efficacy. To improve reproducibility in imaging analysis, a code-free machine learning workflow was developed in Orange Data Mining to classify meibography images. The resulting model classified the severity of gland dropout, demonstrating the potential of modular and user-friendly artificial intelligence platforms for clinical application. In conclusion, this thesis advances understanding of MGD by linking evidence synthesis, clinical practice patterns, therapeutic outcomes and diagnostic innovation. The findings support the integration of thermomechanical therapy into management options and demonstrate the feasibility of automated meibography analysis, providing a foundation for improved standardisation of MGD diagnosis and treatment.