Real-Time Simulation and Visualization of Subject-Specific 3D Lung Dynamics

In this paper we discuss a framework for modeling the 3D lung dynamics of normal and diseased human subjects and visualizing them using an augmented reality (AR) based environment. The framework is based on the results obtained from pulmonary function tests and lung image-data of human subjects obtained from 4D high-resolution computed tomography (HRCT). The components of the framework include a parameterized pressure-volume (PV) relation estimated from normal human subjects, and a physics and physiology-based 3D deformable lung model extracted from the 4D HRCT data of normal and tumor-influenced human subjects. The parameterized PV relation allows modeling different breathing conditions of a human subject. The 3D deformable lung model allows visualizing the 3D shape changes of the lung for the breathing condition simulated by the PV relation. Additionally, the 3D lung model is deformed using a graphics processing unit (GPU) and its vertex shaders, which satisfies the real-time frame-rate requirements of the AR environment