Routine biopsy for managing patients with neurological diseases such as multiple sclerosis (MS), brain tumors, Alzheimer's disease, etc., is not a viable option. Quantitative magnetic resonance imaging (MRI) and magnetic resonance spectroscopic imaging (SI) noninvasively provide tissue morphometric and biochemical information, respectively. This information would be extremely valuable for an objective characterization of the disease state and would greatly help in patient management. However, at present, this powerful technology is not exploited to its fullest extent, except in highly specialized centers, since software that can be routinely used in a typical clinical setting does not exist. The purpose of this proposal is to develop an automatic quantitative MR analysis software package which is robust, user- friendly, and computationally efficient so that it can be routinely used in a typical clinical setting. The proposed software package will allow volume quantitation of white matter (WM), gray matter (GM), cerebrospinal fluid (CSF), lesions, tumors, edema, etc., through MRI and distributions of brain metabolites through SI. This software package will also provide an interface between MRI and SI for an anatomic- metabolic correlation. Such a correlation can be thought of as a "noninvasive biopsy." We will confine ourselves to the development of MRI analysis software during phase I. The necessary software will be developed on a workstation and tested on brain-stimulating phantoms. In phase II, the system will be refined for complete automation, and evaluated on human volunteers. In phase II, the development and evaluation of the software package for SI analysis will also be undertaken.
Thesaurus Terms:computer assisted diagnosis, computer program /software, method development, nervous system disorder diagnosis bioengineering /biomedical engineering, biomedical engineering, clinical biomedical equipmentNational Institute of Neurological Disorders and Stroke (NINDS)
