Font Size: a A A

Optical tomographic imaging using the finite element method

Posted on:2000-04-03Degree:Ph.DType:Dissertation
University:Polytechnic UniversityCandidate:Pei, YalingFull Text:PDF
GTID:1468390014965676Subject:Engineering
Abstract/Summary:
This dissertation is devoted to studies of optical tomographic imaging and more concerns are paid to exploration of measurement strategies to optimize data collection and development of reconstruction algorithms to have a great stability and accuracy on the recovery images with less requirement on measurement techniques and mathematical modeling. The major accomplishments in the Ph.D. study are summarized in the following.; First, a forward model and the corresponding modeling methodologies of tissue structures derived from MRI maps have been developed using an adaptive finite element method to solve diffusion equation. The developed code allows to effectively and easily implement various forward calculations such as the intensity and phase shift at any point on the surface of an object ( i.e., prediction of measurement data) or the distribution of the intensity and phase shift inside an object, for an arbitrary geometry and composition of optical properties.; A systematic study of parameters associated with measurement sensitivity and resolution on homogeneous and inhomogeneous media has been undertaken, based on the developed forward model and solver. The results obtained not only provide a comprehensive understanding on the measurement performance associated with two types of parameter space including measurement and target, but also provide a guide to rational instrument development, and serve as a basis of comparison to clinical studies and aid in test validation.; A very important portion in this dissertation is development of a novel inverse formulation based on a modified perturbation equation using relative detector values. Comparison to the standard perturbation method shows that the new formulation takes substantially less effort to calibration of measurement data, effectively avoids mismatch between measurement system and computational model, greatly reduces the ill-conditioning of optical inverse problem, and has a significant stability and improved accuracy and resolution for imaging reconstruction over a very broad range of optical properties of the selected reference medium. Obviously, these features are very useful for bringing the technique to practical applications.; Specially, the developed forward and inverse codes have been integrated appropriately. The entire program has been written using C++ programming language with a better platform for further development of additional components associated with optical tomographic imaging.
Keywords/Search Tags:Optical tomographic imaging, Using, Measurement, Development
Related items