Agreement between different input image types in brain atrophy measurement in multiple sclerosis using SIENAX and SIENA

Abstract

Purpose: To investigate whether multiple sclerosis (MS) atrophy can be assessed by SIENA and SIENAX software using other image types from MS research protocols than T1-weighted images without contrast agent, which are not always available.

Materials and methods: We selected 46 MS patients with identical magnetic resonance imaging (MRI) protocols at two timepoints. We calculated normalized brain volume (NBV) using SIENAX, and percentage brain volume change (PBVC) using SIENA, from T1-weighted images with and without contrast agent, T2-weighted images, and (calculated) pseudo-T1-weighted images. Relative agreement of the results was assessed using variance component estimation.

Results: Relative agreement with T1-weighted images without contrast agent was good for T1-weighted images with contrast agent (ICC=0.86 for NBV, ICC=0.77 for PBVC), and reasonably good for pseudo-T1 and T2-weighted images (T2: ICC=0.72 for NBV, 0.58 for PBVC; pseudo-T1: ICC=0.68 for NBV, 0.83 for PBVC).

Conclusion: Brain atrophy can be studied using SIENA and SIENAX if T1-weighted images without contrast agent are not available. T1-weighted images with contrast agent should be used if available. Otherwise, pseudo-T1 and T2-weighted images seem acceptable and accessible alternatives. The use of these other images will greatly improve research possibilities, especially regarding older datasets.