[Frontiers in Bioscience 9, 665-683, January 1, 2004]


Robert Zivadinov 1-3, and Rohit Bakshi 1-4

1 Buffalo Neuroimaging Analysis Center, 2 The Jacobs Neurological Institute, 3 Department of Neurology, School of Medicine and Biomedical Sciences, University at Buffalo, State University of New York, 4 Physicians Imaging Centers Buffalo, NY, USA


Figure 1. a: Axial T2-PD weighted image in a 44 year-old woman with SP MS showing multiple hyperintense lesions in the periventricular white matter (arrows). The lesion are characteristic for MS including a size of generally >5mm, oval/ovoid morphology and that many directly contact the ventricular ependyma; b: Sagittal FLAIR image in a 44 year-old man with RR MS showing multiple pericallosal lesions (arrows) with a classic perivenular orientation (Dawson's fingers); c/d: Axial T2-WI of a 33 year-old man with RR MS demonstrating typical hyperintense lesions in the pons (c, arrow) and left cerebellum (d, arrow).

Figure 2. Axial FSE T2-WI (a,c,e) and FLAIR images (b,d,f) in a 27-year old woman with RR MS show typical hyperintense periventricular lesions (a-d) and also involvement of the cortical/juxtacortical region (e,f). FLAIR is superior to T2-WI in the detection of periventricular (b,d) and cortical/juxtacortical lesions (f) because of the heavier T2 weighting combined with the suppression of CSF.

Figure 3. Axial FSE T2-WI (a) and PD-WI (b) in a 42-year old woman with RR MS show typical hyperintense confluent periventricular lesions. Note the improved lesion contrast achieved with PD-WI because of the relatively lower signal intensity of CSF vs. lesions.

Figure 4. Spinal FSE T2-WI of a 28 year old man with RR MS. Sagittal (a) and axial (c) T2-WI show a hyperintense lesion (arrows) in cervical spinal cord (C3-4 level). The lesion is classic for MS, involving less than one spinal level and less than one-half of the cord diameter. The lesion is isointense on noncontrast T1-WI (b), as are most spinal cord MS lesions.

Figure 5. Definition of confluent T2 hyerintense lesions. Figures a and b show single T2 lesions with size larger than 20 mm typically located in the periventricular region and/or on the edges of the anterior and posterior horns of the bodies of ventricles. Figure c and d show areas of white matter abnormalities consisting of two or more T2 lesions interconnected by at least one or more margins.

Figure 6. Semiautomated method of determining T2 hyperintense lesion volume on fast FLAIR images at the Buffalo Neuroimaging Analysis Center in a patient with MS (71). Axial fast FLAIR images show hyperintense lesions on raw images (a-d) and images containing lesions only after completing the algorithm (e-h). Extracranial tissue is first removed using a masking tool that involves an automated contour tracing tool. A threshold technique is applied to separate hyperintense lesions from nonlesional tissue. The software automatically calculates the area and volume of the lesions and total brain lesion volume based on the number of voxels retained.

Figure 7. Single dose (0.1 mmol/kg) gadolinium postcontrast axial T1-WI (a-c) show homogeneously enhancing (hyperintense) lesions (arrows) indicating disruption of the blood-brain barrier and acute inflammation in a 34 year-old woman with RR MS. Double dose (0.2 mmol/kg) gadolinium postcontrast axial T1-WI (d-f) more robustly show enhancing lesions in a 28 year-old man with RR MS.

Figure 8. Axial postcontrast T1-WI demonstrates concentric ring-enhancing lesions (arrows) in the infratentorial (a) and periventricular (b) regions in a 37 year-old woman with SP MS. Characteristic open-ring enhancing lesion with increased specificity for MS (arrow) in 46 year-old man with SP MS (c).

Figure 9. Axial noncontrast T1-WI in a 44 year-old man with SP MS with severe disability (EDSS score 7.0) and multiple hypointense lesions (black holes) in the periventricular white matter on raw images (a and c). Hypointense T1 lesions are segmented for volumetric analysis using an edge finding computerized analysis technique at the Buffalo Neuroimaging Analysis Center (b and d).

Figure 10. Axial magnetization transfer (MT) images of the brain obtained from 2D spin-echo pulse PD-weighted images (PD-WI) at the Buffalo Neuroimaging Analysis Center in a 35 year-old man with SP MS. The image is shown with (a) and without (b) the on-resonance MT saturation radio frequency pulse. The PD-weighted images are co-registered and MT ratio (MTR) maps are created after nulling the CSF and the lesions (c).

Figure 11. Example of MRI diffusion weighted images (DWI) from the Buffalo Neuroimaging Analysis Center in a 44 year-old man with SP MS. Echoplanar DWI is performed generating trace images with a b-factor of 1000 (a) and 0 (b) s/mm2 applied in three orthogonal directions. On the b-1000 DWI map (a) lesions appear as hyperintense areas compared with the surrounding tissue most likely due to T2 shinethrough rather than restricted diffusion. To calculate apparent diffusion coefficients (ADCs), an ADC map (c) is generated from images a and b. The diffusivity was computed separately in the x, y, and z orthogonal directions, and the results averaged to form the mean ADC map (c). Note the hyperintensity of the lesions on the ADC map (c), consistent with T2 shinethrough.