Index of content:
Volume 27, Issue 8, August 2000
- PH. D. THESES ABSTRACTS
Magnetic resonance imaging based correction and reconstruction of positron emission tomography images (in English)27(2000); http://dx.doi.org/10.1118/1.1287650View Description Hide Description
Reconstruction of positron emission tomography(PET)images with good quantitative and qualitative resolution is a difficult problem. Mathematically the problem is ill-posed, and a regularization of the reconstructionsolution is necessary. Such constraints range from simply smoothing the data to the use of elaborate models employed within a Bayesian framework. The goal of optimum resolution, however, is typically not compatible with any of these schemes. This thesis presents new methods for dealing with the projection data on an anisotropic basis. This is done in order to compromise the need for regularization and the drive toward better resolutions. An appropriate resolution is chosen in accordance to either the quality of the projection data (based on local estimates of noise), or known regions of interest (taken, for example, from associated anatomical images). As such it addresses the issues of noise artifact whilst avoiding any unnecessary loss in resolution. A second key element is the development of a Bayesian algorithm to improve resolution in order to address the so-called partial volume effect. Magnetic resonance imaging(MRI) data provides significant information about the expected distribution of radioactivity in the brain. Using this explicit information, corrections to the PETimage can be made in a quantitative manner, but one must be cautious in its application. The thesis describes how to incorporate MRI data into the reconstruction process of the PETimage in an adaptive manner. The adaptational nature of the process is driven solely by good heuristics designed to avoid both an overbiased usage of the a priori information and any assumptions made on the [supposed] homogeneous characteristic of the activity distribution.
27(2000); http://dx.doi.org/10.1118/1.1287651View Description Hide Description
Two new schemes for analysis of mammographicimages were developed, a diagnostic logic for classifying masses on mammograms and a multistage pendulum filter for detecting spicules around the mass. To construct a computer-aided diagnosis scheme for detecting breast cancer requires a knowledge of the diagnostic process. We investigated the classification logic for the diagnosis of breast masses on mammograms in cooperation with a radiologist specialized in mammography. We tested our classification logic by use of 99 mammograms in which the diagnosis had been confirmed. The accuracy for classifying the mass as malignant or benign was very high (sensitivity of 84% and specificity of 96%). Existence of spicules around a mass is one of the important signs which characterize malignant tumors. We developed an automated method for detecting spicules. The detection was performed with a newly developed “multistage pendulum filter.” A “spicule value,” which indicates the probability of “spicule presence” around the mass, was calculated. This filter was evaluated on 71 digitized mammograms. Digitized mammograms were obtained with a laser film scanner with a 0.1 mm×0.1 mm pixel size and 1024 gray levels. The digitized data were then reduced to image with effective pixel size of 0.2 mm for detecting spicules. It detected most of the spicules seen by a radiologist. The sensitivity was 89% and the specificity was 80%, which demonstrates the effectiveness of our method.