DWI scrolling artery sign for the diagnosis of giant cell arteritis: a pattern recognition approach

Area of Science:

• Radiology
• Neuroimaging
• Vascular Imaging

Background:

• Giant cell arteritis (GCA) is a vasculitis affecting large arteries, often involving the head and neck.
• Accurate and timely diagnosis of GCA is crucial to prevent complications like vision loss.
• Current diagnostic methods may have limitations, necessitating novel imaging approaches.

Purpose of the Study:

• To evaluate the diagnostic accuracy of a pattern recognition method on head MRI using diffusion-weighted imaging (DWI) for suspected GCA.
• To assess the reliability and speed of the DWI-Scrolling-Artery-Sign (DSAS) for GCA diagnosis.

Main Methods:

• Retrospective analysis of 156 patients with suspected GCA undergoing head MRI.
• Identification and rating of the DWI-Scrolling-Artery-Sign (DSAS) by experts and a novice.
• Comparison of DSAS findings with T1-weighted black-blood sequences and clinical diagnosis after follow-up.

Main Results:

• The DSAS demonstrated high sensitivity and specificity in diagnosing GCA, with expert performance at 73.6% sensitivity and 94.2% specificity.
• Excellent inter-reader agreement was observed for DSAS assessment among experts and between experts and a novice.
• DSAS assessment was rapid, taking less than one minute per scan.

Conclusions:

• The DWI-Scrolling-Artery-Sign (DSAS) is a reliable imaging biomarker for GCA diagnosis.
• This pattern recognition approach using DWI offers good diagnostic accuracy and can be implemented quickly in clinical practice.