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

Abstract

Herpes zoster (HZ) can cause a blistering skin rash with severe neuropathic pain. Pharmacotherapy is the most common treatment for HZ patients. However, most patients are usually the elderly or those that are immunocompromised, and thus often suffer from side effects or easily get intractable post-herpetic neuralgia (PHN) if medication fails. It is challenging for clinicians to tailor treatment to patients, due to the lack of prognosis information on the neurological pathogenesis that underlies HZ. In the current study, we aimed at characterizing the brain structural pattern of HZ before treatment with medication that could help predict medication responses. High-resolution structural magnetic resonance imaging (MRI) scans of 14 right-handed HZ patients (aged 61.0 ± 7.0, 8 males) with poor response and 15 (aged 62.6 ± 8.3, 5 males) age- ( p = 0.58), gender-matched ( p = 0.20) patients responding well, were acquired and analyzed. Multivoxel pattern analysis (MVPA) with a searchlight algorithm and support vector machine (SVM), was applied to identify the spatial pattern of the gray matter (GM) volume, with high predicting accuracy. The predictive regions, with an accuracy higher than 79%, were located within the cerebellum, posterior insular cortex (pIC), middle and orbital frontal lobes (mFC and OFC), anterior and middle cingulum (ACC and MCC), precuneus (PCu) and cuneus. Among these regions, mFC, pIC and MCC displayed significant increases of GM volumes in patients with poor response, compared to those with a good response. The combination of sMRI and MVPA might be a useful tool to explore the neuroanatomical imaging biomarkers of HZ-related pain associated with medication responses.

Abnormal Intrinsic Brain Activity and Neuroimaging-Based fMRI Classification in Patients With Herpes Zoster and Postherpetic Neuralgia

“…The SVM classifier performed between HZ and PHN patients had a classification accuracy of 61.54%. We did not achieve as high a classification accuracy as in previous studies (31,46). It should be noted that, in our study, all patients have taken pregabalin and amitriptyline to alleviate the pain intensity before the MRI scan.…”

“…For patients with neuropathic pain, one previous neuroimaging study has found that the voxelmirrored homotopic connectivity in the dorsolateral prefrontal cortex, precuneus, and PCC discriminated between patients with PHN and healthy subjects (31). Our recent study measuring gray matter volume found that alterations in several brain regions, including the middle frontal cortex, ACC, precuneus, and cuneus, had a significant predictive power to classify HZ patients with different responses to medications (46). Along with these findings, our observations suggested that the frontal gyrus and the precuneus have a significant predictive power to distinguish neuropathic pain patients from the healthy controls.…”

Stage detection of mild cognitive impairment via fMRI using Hilbert Huang transform based classification framework

“…Since Cortes and Vapnik proposed the Support Vector Machine (SVM) technique, 29 it has been widely used in brain science research to solve the binary classification problem 27,30–38 . The decision function of SVM is shown as follows:where Ƙ represents the kernel function.…”

The Spontaneous Activity Pattern of the Middle Occipital Gyrus Predicts the Clinical Efficacy of Acupuncture Treatment for Migraine Without Aura

“…Initially, the inputted zALFF maps were masked with an Anatomical Automatic Labeling 116 gray matter atlas to remove the signals of white matter and cerebrospinal fluid and reduce the computation burdens. We combined the searchlight algorithm (42) and principal component analysis (PCA) to conduct feature selection referencing the previous studies (43)(44)(45)(46). Briefly, at each voxel (Vi), a spherical cluster with a 9-mm radius was defined centering at Vi, and the value of each voxel within the spherical cluster was extracted to generate the feature matrices.…”

“…The meaningful classifying features were defined as those clusters with a minimum accuracy of 70% and more than 50 contiguous voxels. Same to the previous studies (44,45), the accuracy of each cluster was determined by its peak accuracy.…”