Related Experiment Video
Updated: Jan 12, 2026
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Utility of Locus Coeruleus Signal Intensity on High-resolution T1-weighted MR Imaging with Magnetization Transfer for
Yuta Yoshimatsu1, Satoru Ide1, Naoki Ogawa1
1Department of Radiology, University of Occupational and Environmental Health School of Medicine, Kitakyushu, Fukuoka, Japan.
Purpose:
This study aimed to evaluate the diagnostic utility of locus coeruleus (LC) signal intensity on high-resolution T1-weighted imaging with magnetization transfer (T1WI with MT) for distinguishing Parkinson's disease (PD) from a broad range of atypical parkinsonism (AP) subtypes, including early-stage cases.
Methods:
We retrospectively analyzed T1WI with MT data from 214 participants, including patients with PD (n = 125), corticobasal syndrome (CBS, n = 12), multiple system atrophy (MSA, n = 16), progressive supranuclear palsy (PSP, n = 19), essential tremor (n = 17), vascular parkinsonism (n = 4), drug-induced parkinsonism (DIP, n = 7), and healthy subjects (HS, n = 14). Circular ROIs were placed on the LC and substantia nigra pars compacta to calculate contrast ratios (CRs). Conventional MRI findings of AP, focusing on characteristic regional atrophy patterns, were also evaluated. Diagnostic performance was assessed using receiver operating characteristic (ROC) analysis. A subanalysis was performed for early-stage cases (within 2 years of onset). Three independent neuroradiologists evaluated T1WI with MT, and interobserver agreement was assessed using intraclass correlation coefficients (ICC).
Results:
The LC-CR was significantly lower in PD than in HS (P < 0.01) and all AP subtypes except DIP (P = 0.37). ROC analysis revealed that LC-CR had the highest diagnostic accuracy for differentiating PD from AP (area under the curve [AUC] = 0.83, sensitivity = 67%, specificity = 90%). In early-stage cases, LC-CR maintained high specificity (98%) with an AUC of 0.80. The diagnostic utility of LC-CR was comparable or superior to conventional MRI findings in distinguishing PD from CBS, MSA, and PSP. Interobserver agreement for LC-CR measurements was good, with an ICC of 0.87 (95% confidence interval: 0.85-0.89).
Conclusion:
LC-CR measured on T1WI with MT serves as a reliable imaging biomarker for differentiating PD from various forms of AP, even in early disease stages.
