Multiple Modality Imaging

There are two types of multimodal integrative imaging studies that are important in the clinical evaluation of patients. The first is the within-subject integration of information from multiple brain mapping techniques, or the serial integration of multiple imaging studies in time using the same technique in the same individual. The second approach is the averaging or integration of information from multiple subjects, a much more difficult problem due to the great anatomical and functional variability that exists between individuals. The within-subject integration problem has yielded to a variety of excellent and elegant mathematical approaches for the alignment and registration of data. The between-subject problem is a more difficult one that is being resolved through the use of warping and morphing techniques.

1. Within-Subject Registration

A composite image of a patient derived from imaging using multiple imaging modalities or serial studies over time is a critical indicator of the clinical picture from an imaging perspective-for example, in a patient with a brain tumor in whom the natural history of the enlargement of the lesion can be evaluated quantitatively and objectively. Similarly, the integrated image of functional activation in cortical regions surrounding a lesion that is to be surgically resected can predict the relative risk of functional damage due to resection of normal cortex in the process of tumor ablation (Fig. 11).

Images of interictal spikes can be obtained by combining EEG and fMRI. Such images capitalize on the excellent temporal resolution of EEG and the complementary high spatial resolution of MRI. The relationship between hypometabolism in a seizure focus, determined with PET measurements of glucose metabolism, can be compared with benzodiazepine receptor binding, thereby increasing the specificity and

Figure 11 Within-subject registration techniques. (A) Three-dimensional reconstruction of a patient's brain with a cortical tumor in the region ofthe sensorimotor cortex (arrow). The structural data set, reconstructed from MRI, can then be combined with functional information about cerebral blood flow changes associated with motor tasks, derived from 15O-labeled water PET studies when the patient moved the left leg (B), shoulder (C), or fingers (D). (E) Intraoperative view of the hand area of the motor cortex where the localization of sensorimotor hand function was identified intraoperatively with electrophysiologic techniques and labeled "1" and "2." (F) Operative view of site following resection of tumor. Because of the close proximity of the tumor to the activation site for finger movement, identified with preoperative PET imaging, it was predicted that this patient would have loss of fine motor control of the hand following complete resection of the lesion. This was in fact the case and is indicative of the accuracy and predictive power of preoperative mapping in patients with cortical lesions close to vital cortical structures (courtesy of Roger Woods, ULCA School of Medicine, and Scott Grafton, Emory University School of Medicine).

sensitivity of the combined result. One can also combine electrophysiologic data sets with tomograph-ic data from PET, MRI, and MRS to provide a composite preoperative assessment of patients with epilepsy. PET measurements of cerebral blood flow, oxygen extraction, and oxygen metabolism in the same subject and comparison with diffusion-weighted MRI and MR angiography (or helical CT) provide a very complete picture of the supply-demand relationships of the brain parenchyma in patients with ischemic cerebrovascular disease. Such combined studies will undoubtedly become a clinical norm rather than an exception in the future.

Comparisons of scans between individual patients will become increasingly important. Such comparisons require that the scans be spatially normalized to account for individual differences in brain structure. The ability to normalize a scan to a standard brain space means that individual patient scans can be compared with a representative scan from a population of normal subjects that takes into account a realistic estimate of the anatomical variability in that population. In the structural domain, this ability may increase the sensitivity with which subtle heterotopias or other migrational abnormalities are identified in patients with focal epilepsy. Similarly, selected patterns of atrophy in neurodegenerative diseases should be detected in a more sensitive and specific fashion. The ability to compare representative scans across patient groups could also have importance for clinical trials where a patient group on an experimental therapy could be compared with a control group in an objective and quantifiable manner.

Was this article helpful?

0 0
Stop Anxiety Attacks

Stop Anxiety Attacks

Here's How You Could End Anxiety and Panic Attacks For Good Prevent Anxiety in Your Golden Years Without Harmful Prescription Drugs. If You Give Me 15 minutes, I Will Show You a Breakthrough That Will Change The Way You Think About Anxiety and Panic Attacks Forever! If you are still suffering because your doctor can't help you, here's some great news...!

Get My Free Ebook


Post a comment