The purpose of this proposal is the continued development and evaluation of 3-D Whole Body PET as a tool for the clinic and research. Modern PET systems approach the realistic limit for raw sensitivity by surrounding the body with efficient detector material, and full 3-D data collection. With this high, sensitivity, PET images still have a relatively poor spatial resolution and S/N. Taking full advantage of 3-D PET requires development of the best methods of data collection, data correction, and image reconstruction. Specifically: 1. Development of 3-D data collection, data correction and system calibration techniques and their reduction to practice on new high resolution, long axial field- of- view PET systems. In this grant, new methods of data correction, data collection, and instrument calibration were developed and put into practice. A number of methods are still in development. However, all work was performed with old low -resolution systems. To realize the full potential of 3-D Whole Body PET, it is necessary to implement these methods on true 3-D PET systems, and continue development and validation of unfinished techniques. These include a new Monte Carlo based scatter correction, and a new method of 3-D whole body attenuation correction. 2. Implementation of 3-D reconstruction algorithms in 3-D Whole Body PET and evaluation of their properties. 3-D reconstruction algorithms will be implemented and evaluated for the high resolution PET systems. In addition, 3-D interactive reconstruction techniques will also be implemented and investigated. When implemented appropriately, 2-D iterative reconstruction techniques have been shown to improve image S/N and lesion detection in PET. A similar improvement is anticipated for 3-D PET. The project will develop fast methods of performing both types of 3-D reconstructions, and will quantitatively assess that changes in S/N and lesion detection that can be attributed to 3-D PET and/or the application of the iterative reconstruction techniques . 3. Receiver Operating Characteristic/Location Sensitive Receiver Operating Characteristic (ROC/LROC) study of 2-D and 3-D Whole Body PET. To determine whether there is a significant practical difference among these techniques, when a physician is part of the measuring system, an LROC/ROC study will be carried out to provide objective measures of the ability to detect lesions with 2-D and 3-D and 3-D whole body PET. The ROC/ROC study will use well defined simulated lesions added to the image data of normal subjects before reconstruction. 4. Study of the relative effectiveness of 2-D and 3-D whole body PET in detection of lung cancer. In order to examine the effectiveness of 3-D whole body PET and the various methods of reconstructing the 3-D data sets, a 3-D data collection over the thorax will be added to the diagnostic workup of the 40-50 lung CA patients who undergo surgery at UCLA per year, who are also scheduled for 2-D whole body PET. In summary, implementation of new 3-D PET imaging techniques on high resolution 3-D PET systems is proposed with objective and quantitative testing performed with simulation and phantoms, the subjectivity of the physician measured with ROC, and the clinical usefulness tested by imaging lung cancer patients.
Showing the most recent 10 out of 17 publications
