Method for quantification and modeling of specimens using microscopic image slices
Challenges facing biomedical images analysts in the growing field of microscopic imaging encompass the task of gathering evidentiary data when presented with volume of digital microscopic images, which may possibly contain critical hidden information. This thesis discusses the process and necessary considerations inherent in the development of methods applied for the detection of the hidden information which otherwise may not be available. Well known is the growing trend and the use of high resolution microscope. Whether expensive equipment or user specific microscopes , digital images obtained aid in better understanding of the specimen involved in number of diverse fields such as medicine, biological research, cancer research, drug testing etc. As a result of such, a new image quality enhancement and 3D visualization techniques presented; not only receive derivation from a foundation of general enhancement theory, but also incorporate a thorough understanding of the shortcomings of the microscopes and process of natural digitization of images obtained.
Fundamental to this investigation, an understanding of the use of microscopes in biomedical imaging and the challenges facing analysts in recovering complete information through various image enhancement process. Analyses began with a concentration on accurately and efficiently creating methods in the detection and enhancement of hidden information and also painlessly interfacing with existing 3D visualization software. Notably, due to large volume of these data sets obtained from microscopes, fast, automated and reliable processing of these images is required by the analysts. The research in this thesis attempts to understand the two primary investigation techniques in microscopic imaging. The first goal is to gain understanding of the nature of the microscopes and its defects in providing complete and true information. The second objective is to investigate and develop image quality enhancement techniques that are used to bring out the hidden information within by understanding the nature of images.
In the design portion of the thesis presented is a new and modified image quality enhancement method which aids in bringing out hidden information from within microscopic images. The investigation begins with the basic understanding of microscopes in relation to biomedical imaging. Data sets from two broadly used microscopes: electron microscope (µ-CT scans) and light microscope (two-photon excited fluorescence microscopic images) were obtained. Then further the study of literature regarding several existing image enhancement techniques and approaches that have been developed over the last few years. A common method was created to better suit any microscopic images and enhances these huge data sets that will eventually help in better visualization in 3D. Current limitations of these implementations include the inability of software to process these enhanced image data sets in 3D to obtain required statistical evidence in better understanding of the specimen. And also this new method developed requires high performance processors for faster enhancement and evaluation. Finally, the findings and analysis of the system are evaluated in comparison.