Proton magnetic resonance spectroscopic imaging has been demonstrated to have significant clinical potential for the noninvasive localization brain tumors. In vivo spectroscopic measurements have revealed changes in brain metabolites that can be used to classify tissues as healthy or cancerous. However, it has been shown that radiation therapy changes the metabolite levels in healthy tissue, thus making it more difficult to distinguish between healthy tissue and tumor using spectroscopy alone. To overcome this problem, this research will study the integration of spectroscopy with diffusion and perfusion imaging into an automated tissue classification system. The time- and dose-response of radiation-induced changes in healthy brain tissue will be assessed by analyzing serial spectroscopic, diffusion, and perfusion imaging data from patients undergoing radiation therapy. The abnormality of a given volume of tissue will then be defined as the difference between the measured and expected values for a given parameter. An overall abnormality index will then be computed from a combination of spectroscopic, diffusion, and imaging parameters. It is hypothesized that this will improve the accuracy in identifying tumor in post-radiotherapy examinations compared with conventional imaging or MRSI alone. This will allow clinicians and researchers to better understand patterns of failure following radiotherapy of high grade gliomas, and adjust accordingly.
Lee, Michael C; Nelson, Sarah J (2008) Supervised pattern recognition for the prediction of contrast-enhancement appearance in brain tumors from multivariate magnetic resonance imaging and spectroscopy. Artif Intell Med 43:61-74 |
Lee, Michael C; Cha, Soonmee; Chang, Susan M et al. (2006) Partial-volume model for determining white matter and gray matter cerebral blood volume for analysis of gliomas. J Magn Reson Imaging 23:257-66 |