Abstract WP101: Clinical Utility Of Simultaneous Multislice Diffusion MRI In Acute Stroke Workup

Abstract

Purpose: Rapid and accurate ischemic stroke detection is essential for optimal management. Diffusion tensor imaging (DTI) trace sequences are highly accurate for acute stroke detection but require long acquisition times. This study aimed to assess the impact of simultaneous-multislice (SMS) acceleration on scan time, image quality, and diagnostic accuracy. Methods: In this IRB-approved study, MRI brain imaging was performed during stroke workup on Siemens 1.5T scanners with a 16-channel head coil. DTI (20 directions, b-value of 2000 sec/mm 2 ) was acquired without and with slice acceleration factor 2 (SMS-2). DTI trace-positivity (stroke or other) was independently assessed by two neuroradiologists, blinded to sequence parameters and clinical data. Average image quality (5-point scale, 5=excellent) was measured between the radiologists for each MRI. Differences in signal-to-noise ratio (SNR), coefficient of variation (CV), repetition time (TR), interrater agreement (Kappa), and receiver operating characteristic (ROC) areas were determined using a p<0.05 significance threshold. Results: SMS-2 significantly reduced TR, equivalent to 45.0% reduced scan time. Image quality and SNR were slightly but significantly reduced in SMS-2-DTI compared to standard (STD)-DTI (N=41, see table and figure ). Interrater agreement for DTI-positivity was ‘excellent’ on both STD-DTI and SMS-2-DTI without significant difference. ROC areas of each observer were also not significantly different between STD-DTI and SMS-2-DTI. Conclusion: SMS-2 reduces scan time. Image quality was slightly reduced, however, satisfactory diagnostic accuracy for stroke detection was maintained.