Yoshimi Anzai MD, Presenter: Nothing to Disclose

Gregory James Wilson PhD, Abstract Co-Author: Nothing to Disclose

Kenneth R. Maravilla MD, Abstract Co-Author: Nothing to Disclose

Voxel based morphometric and volumetric brain segmentation has been used to investigate neurodegenerative disease, in particular in patients with memory loss or cognitive decline. This technique is being applied to clinical practice with availability of FDA approved software, such as NeuroQuant (CorTechs Labs Inc. CA). This automated segmentation program requires a 1.2 mm3 voxel, 3D IR-TFE (inversion recovery turbo field echo) sequence that currently takes 9 minutes duration (ADNI sequence), often resulting in patient motion in this cognitively impaired population. We have implemented and tested a shorter IR-TFE sequence (5 minutes) and compared with ADNI recommended sequences.

All MR data was acquired on a 3T wide-bore whole-body scanner. Volumes of 12 automatically segmented (NeuroQuant) brain structures in each side of brain (total of 24 structures per subject), asymmetric index (R vs. L) of the 12 brain structures, and age adjusted percentile of hippocampal volume were recorded in both 9 min and 5 min IR-TFE sequences in four subjects. Both MR acquisition protocols are exactly the same except for application of SENSE factor 2 in R-L direction and an increase in slice oversampling factor from 1.28 to 1.4 for 5 min scan. Spearman’s correlation was used for the analysis of asymmetric index and actual volume measurements of the anatomic brain structures.

The volume of brain structures was very similar between 9 min and 5 min scans. Average difference in brain volume of brain structures was 0.32 cc, ranging from 0.0 cc to 2.34 cc. Spearman’s nonparametric correlation shows extremely high correlation for brain volumetric (ρ=0.9977) as well as asymmetric index (ρ =0.9257). (Graph 1).

By applying SENSE factor of 2 and increasing the slice-oversampling factor by 10%, we were able to reduce the scan time of brain volumetric 3D IR-TFE sequence by 45% from 9 min to 5 min. This can be easily implemented into clinical practice for application in memory-impaired patients who might benefit from quantitative brain imaging.

Volumetric brain analysis can provide valuable information for characterizing neurodegenerative disease. We implemented a shortened sequence without compromising accuracy of volumetric analysis.

Anzai, Y, Wilson, G, Maravilla, K, Effectiveness of Reduced MR Scan Time for Volumetric Brain Analysis.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14016599.html