Design and development of drug-induced sleep endoscopy classification system, data fusion system and relational data model

Abstract

There is a lack of consensus on how to assess and report upper airway obstruction based drug-induced sleep endoscopy findings, a lack of multiple information visualisation system for real-time drug-induced sleep endoscopy assessment and a lack of a relational model for drug-induced sleep endoscopy information management. The key objective of this thesis is to design and develop drug-induced sleep endoscopy data management systems. An integrated drug-induced sleep endoscopy classification model was developed to act as a referential model based on a systematic literature review conducted on drug-induced sleep endoscopy classification systems. A clinical feasibility study using drug-induced sleep endoscopy data fusion system was conducted for proof of concept validation based on 20 drug-induced sleep endoscopy assessments. A relational drug-induced sleep endoscopy model for drug-induced sleep endoscopy database information management was developed by reviewing 30 clinical folders of sleep-disordered breathing patients). Integrated drug-induced sleep endoscopy classification model was developed as an integrated and referential drug-induced sleep endoscopy classification system for four commonly used drug-induced sleep endoscopy classification systems. Drug-induced sleep endoscopy data fusion system was used to capture, merge, display and store upper airway obstruction events and its associated cardiorespiratory changes during drug-induced sleep endoscopy assessment for 20 sleep-disordered breathing patients. A conceptual and physical model of a relational drug-induced sleep endoscopy model for drug-induced sleep endoscopy database management system was developed consisting of 8 entities and 7 relationships. A referential drug-induced sleep endoscopy classification model provides a framework for comparing multiple drug-induced sleep endoscopy findings. A multiple information visualisation system provides critical information on anatomical severity and its associated cardiorespiratory changes during drug-induced sleep endoscopy assessment in real-time. A relational data model for drug-induced sleep endoscopy database management system provides a framework towards developing the first web drug-induced sleep endoscopy information management system and drug-induced sleep endoscopy informatics system.