Computational Modeling of Optical Projection Tomographic Microscopy

Abstract

Lung cancer is the leading cause of cancer related deaths worldwide with 1.37 millions deaths per year. In 2011, Computed Tomography became the first method approved for early detection, however it is hindered by safety concerns due to its use of radiation and it is only capable of detecting masses greater than 2mm in diameter. In an effort to perform early detection before masses are visible, VisionGate Inc. is developing a microscope based on Optical Projection Tomographic Microscopy (OPTM) to detect cancer at the cellular level. OPTM reconstructs three-dimensional images of single cells from projection images acquired from different perspectives around the cell. This dissertation presents a computer simulation that is designed to analyze all aspects of OPTM image formation, such as degree of specimen staining, refractive index matching, and objective scanning. The computer simulation is used in conjunction with experimental methods to refine conclusions and improve system design in a robust methodology. While this work is currently driving improved microscope design, in the long term it will provide a framework to develop quantitative microscopy where pixel values are directly related to cellular optical properties. These improvements will provide improved sensitivity and specificity for early detection and ultimately reduce lung cancer mortality.