Salvage Radiosurgery for Brain Metastases: Prolonged Survival With Durable Tumor Control
| Autor: | Genevieve Gingras-Hill, Gelareh Zadeh, B.A. Millar, Caroline Chung, Cynthia Ménard, Mark Bernstein, N.J. Laperriere, G. Kurtz |
|---|---|
| Rok vydání: | 2012 |
| Předmět: |
Cancer Research
medicine.medical_specialty Radiation Orientation (computer vision) Image quality business.industry medicine.medical_treatment Magnetostatics Tumor control Magnetic susceptibility Imaging phantom Radiosurgery Nuclear magnetic resonance Oncology Signal-to-noise ratio (imaging) medicine Radiology Nuclear Medicine and imaging Medical physics business |
| Zdroj: | International Journal of Radiation Oncology*Biology*Physics. 84:S825 |
| ISSN: | 0360-3016 |
| DOI: | 10.1016/j.ijrobp.2012.07.2210 |
| Popis: | metastases and suggests practical solutions to mitigate their undesired effects. Materials/Methods: Serial patients treated with radiosurgery with hemorrhagic brain metastases well-defined on CT were included in this study. For each patient, CT and corresponding Gd-enhanced T1-weighted FSPGR, acquired on a 3T MRI, were co-registered using a commercial rigid mutual information algorithm. Subsequently, the target lesions were manually segmented on the MR images. The scanner-related distortions (i.e., B0 inhomogeneity and gradient non-linearities) were measured by using a dedicated phantom. The susceptibility geometric distortions (Dx) were modeled by numerically solving the Maxwell equations for a static magnetic field, i.e., div[(1+ChiM)grad(PhiM)] Z 0, where PhiM is the magnetic scalar potential and ChiM is the magnetic susceptibility constant. The input data required by the simulations consisted of the spatial distribution of susceptibility values in the patient anatomy (i.e., soft-tissue, air, bone, hemorrhagic mets) derived from CT and PhiM boundary values. The simulations were performed for B0 of 0.2T to 5T and GE of 2 to 20 mT/m. The CT contours were modified to include the simulated susceptibility distortions and then compared to the MR contours. Results: The magnitude of the susceptibility effects for hemorrhagic metastases depends on several parameters including the size, shape and iron content of the metastasis. There is a strong dependence on the strength of B0 and encoding gradient (GE) (i.e., Dx w B0/GE), and a weak influence due to the relative orientation of B0 with regard to the volume shape. The geometric distortions amounted to about 1.5 mm for B0 and GE values of 3T and 4 mT/m, respectively. The MR acquisition may be optimized by increasing GE strength (for a given B0) to reduce the distortions affecting the target volumes such that residual distortions are not clinically significant. However, the increase of GE introduces an image quality penalty due to the decrease in the signal to noise ratio. Conclusions: The susceptibility effects were investigated and quantified for several parameters. The correction of only the system-related distortions may not be enough for an accurate MR-based target definition. The susceptibility effects need to be carefully investigated and their impact quantified. Author Disclosure: T. Stanescu: None. C. Chung: None. A. Simeonov: None. M. van Prooijen: None. B. Millar: None. C. Menard: None. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|