Characterizing the Structural Pattern Predicting Medication Response in Herpes Zoster Patients Using Multivoxel Pattern Analysis

The development of high-resolution neuroimaging and multielectrode electrophysiological recording provides neuroscientists with huge amounts of multivariate data. The complexity of the data creates a need for statistical summary, but the local averaging standardly applied to this end may obscure the effects of greatest neuroscientific interest. In neuroimaging, for example, brain mapping analysis has focused on the discovery of activation, i.e., of extended brain regions whose average activity changes across experimental conditions. Here we propose to ask a more general question of the data: Where in the brain does the activity pattern contain information about the experimental condition? To address this question, we propose scanning the imaged volume with a "searchlight," whose contents are analyzed multivariately at each location in the brain.

Brain responses to pain, assessed through positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) are reviewed. Functional activation of brain regions are thought to be reflected by increases in the regional cerebral blood flow (rCBF) in PET studies, and in the blood oxygen level dependent (BOLD) signal in fMRI. rCBF increases to noxious stimuli are almost constantly observed in second somatic (SII) and insular regions, and in the anterior cingulate cortex (ACC), and with slightly less consistency in the contralateral thalamus and the primary somatic area (SI). Activation of the lateral thalamus, SI, SII and insula are thought to be related to the sensory-discriminative aspects of pain processing. SI is activated in roughly half of the studies, and the probability of obtaining SI activation appears related to the total amount of body surface stimulated (spatial summation) and probably also by temporal summation and attention to the stimulus. In a number of studies, the thalamic response was bilateral, probably reflecting generalised arousal in reaction to pain. ACC does not seem to be involved in coding stimulus intensity or location but appears to participate in both the affective and attentional concomitants of pain sensation, as well as in response selection. ACC subdivisions activated by painful stimuli partially overlap those activated in orienting and target detection tasks, but are distinct from those activated in tests involving sustained attention (Stroop, etc.). In addition to ACC, increased blood flow in the posterior parietal and prefrontal cortices is thought to reflect attentional and memory networks activated by noxious stimulation. Less noted but frequent activation concerns motor-related areas such as the striatum, cerebellum and supplementary motor area, as well as regions involved in pain control such as the periaqueductal grey. In patients, chronic spontaneous pain is associated with decreased resting rCBF in contralateral thalamus, which may be reverted by analgesic procedures. Abnormal pain evoked by innocuous stimuli (allodynia) has been associated with amplification of the thalamic, insular and SII responses, concomitant to a paradoxical CBF decrease in ACC. It is argued that imaging studies of allodynia should be encouraged in order to understand central reorganisations leading to abnormal cortical pain processing. A number of brain areas activated by acute pain, particularly the thalamus and anterior cingulate, also show increases in rCBF during analgesic procedures. Taken together, these data suggest that hemodynamic responses to pain reflect simultaneously the sensory, cognitive and affective dimensions of pain, and that the same structure may both respond to pain and participate in pain control. The precise biochemical nature of these mechanisms remains to be investigated.

An overview of computational procedures for examining neuroanatomical variability is presented. The review focuses on approaches that can be applied using the SPM software package, beginning by explaining briefly how statistical parametric mapping is usually applied to functional imaging data. The review then proceeds to discuss volumetry, with an emphasis on voxel-based morphometry, and the pre-processing steps involved using the SPM software. Most volumetric studies involve univariate approaches, with a correction for some global measure, such as total brain volume. In contrast, the overall form of the brain may be more accurately modeled using multivariate approaches. Such models of anatomical variability may prove accurate enough for computer assisted diagnoses.