A decision tree for differentiating multiple system atrophy from Parkinson's disease using 3-T MR imaging

Area of Science:

• Neuroimaging
• Neurology
• Radiology

Background:

• Differentiating multiple system atrophy (MSA) from Parkinson's disease (PD) is clinically challenging.
• Accurate diagnosis is crucial for appropriate treatment and management of these neurodegenerative disorders.
• Conventional MRI and diffusion tensor imaging (DTI) offer potential biomarkers for differential diagnosis.

Purpose of the Study:

• To develop and validate a decision tree model for distinguishing MSA from PD.
• To utilize standard magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI) parameters.
• To achieve high diagnostic accuracy in differentiating these two conditions.

Main Methods:

• 3-Tesla brain MRI and DTI were acquired from 26 PD and 13 MSA patients.
• Regions of interest included the putamen, substantia nigra, pons, middle cerebellar peduncles (MCP), and cerebellum.
• Measurements included volumetric data and DTI metrics (fractional anisotropy, mean diffusivity); a three-node decision tree was constructed.

Main Results:

• Nine imaging parameters showed significant differences between MSA and PD (P < 0.05).
• The decision tree incorporated mean MCP width, pons anteroposterior diameter, and mean MCP FA.
• The decision tree achieved 92% sensitivity, 96% specificity, 92% positive predictive value, and 96% negative predictive value, correctly classifying 12 of 13 MSA patients.

Conclusions:

• A decision tree model based on MRI and DTI parameters can effectively differentiate MSA from PD.
• This imaging-based approach provides a descriptive and predictive tool for diagnosis.
• Combined conventional MRI and DTI enhance diagnostic accuracy for differentiating these parkinsonian syndromes.