Computational mechanics models for studying the pathogenesis of obstructive sleep apnea (OSA)

The goal of this thesis is to investigate new computational tools for understanding normal upper airway mechanics and OSAS pathogenesis, a sleep disorder in which the complete occlusion of the upper airway accompanies the cessation of airflow repeatedly during sleep.; Firstly 3D patient-specific computational fluid dynamics (CFD) models were developed to investigate the contributions of airway anatomical restrictions to the airflow and resistance in children with OSAS and matched control subjects. Validated by in-vitro experiments, normal controls had significantly less pressure drop in the pharynx than the nasal passages, but in the OSAS cases maximum pressure drop in the nasopharynx ranged from 30% to 1700% higher than nasal pressure drop. These new findings suggest that the shape of pharynx is an important factor in driving internal pressure toward the collapse pressure. Simplification and verification were discussed. Pressure distribution and resistance in pharynx have been shown strongly correlated with the airway cross-section area profile, and the accuracy is also influenced by the length of the narrowed segment and the pressure recovery coefficient.; Contributions of anatomical restriction on airway collapse are investigated using a two-dimensional fluid structure interaction (FSI) model. The model couples internal flow with upper airway mechanics, and reveals that airway narrowing could be also an important factor determining airway patency, besides clinical measurements Pcrit. The effects of upper airway muscle activation in response to negative pharyngeal pressure to maintain airway patency are impaired by the anatomical restriction.; To study the structure and tissue properties non-invasively, a modeling method was developed that extends published methods to determine material properties of passive diastolic myocardium. MR imaging, MRI tissue tagging, finite element analysis, and nonlinear optimization, are used to identify model structure and tissue properties of the deformable upper airway. The model incorporates airway architecture and intrinsic material properties, and has been validated by noninvasive MR tagging techniques. Baseline studies demonstrate correct qualitative response, and quantitative accuracy of the model displacements. A parameter sensitivity study indicates that the airway collapsibility is most sensitive to the tongue mechanical property.