Identifying the presence of Parkinson's disease using low-frequency fluctuations in BOLD signals. (3rd April 2017)

• Record Type: Journal Article
• Authors: Tang, Yan; Meng, Li; Wan, Chang-min; Liu, Zhen-hua; Liao, Wei-hua; Yan, Xin-xiang; Wang, Xiao-yu; Tang, Bei-sha; Guo, Ji-feng
• Is Part Of: Neuroscience letters. Volume 645(2017)
• Page Start: 1
• Page End: 6
• Publication Date: 2017-04-03
• Subjects: Parkinson's disease -- Resting-state functional magnetic resonance imaging -- Amplitude of low-frequency fluctuations -- Fractional amplitude of low-frequency fluctuations; Neurology -- Periodicals; Neurology -- Periodicals; Research -- Periodicals; Neurologie -- Périodiques; Neuroanatomie -- Périodiques; Neuropharmacologie -- Périodiques; Neurophysiologie -- Périodiques; Neurology; Periodicals; Electronic journals; 617.48
• DOI: 10.1016/j.neulet.2017.02.056
• Languages: English
• ISSNs: 0304-3940

Abstract

Highlights: Using the leave-one-out cross-validation method, PD can be reliably distinguished from the controls with 92% sensitivity and 87% specificity. Support vector machines (SVM)-neuroimaging approach may be of particular clinical value. RS-fMRI should be considered for development as a biomarker and an analytical tool for the evaluation of PD. Abstract: Parkinson's disease (PD) is a chronic, progressive, and degenerative neurological disorder that is characterized by the degeneration of dopamine neurons in the substantia nigra and the formation of intracellular Lewy inclusion bodies. Resting-state functional magnetic resonance imaging (RS-fMRI) has demonstrated evidence of changes in metabolic patterns in individuals with PD. The purpose of this study was to determine whether the presence of PD could be "predicted" based on resting fluctuations in the blood oxygenation level dependent signal. We utilized RS-fMRI to measure the amplitude of low-frequency fluctuation (ALFF) and the fractional ALFF (fALFF) in 51 patients with PD and 50 age- and sex-matched healthy controls. Compared with the healthy controls, the individuals with PD exhibited altered ALFFs in the bilateral lingual gyrus and left putamen and an altered fALFF in the right cerebellum posterior lobe. Support vector machines (SVMs), which comprise a supervised pattern recognition method that enables predictions at the individual level, were trained to separate individuals with PD from healthy controls based on the ALFF and fALFF. Using the leave-one-out cross-validation method to analyze our sample, we reliably distinguished the participants with PD from the controls with 92% sensitivity and 87% specificity. Overall, these findings suggest that the SVM-neuroimaging approach may be of particular clinical value because it enables the accurate identification of PD at the individual level. RS-fMRI should be considered for development as a biomarker and an analytical tool for the evaluation of PD.