A Processing Pipeline for Quantifying Lenticulostriate Artery Vascular Volume in Subcortical Nuclei

Abstract

Lenticulostriate arteries (LSAs) supply blood to the basal ganglia region. Its lesion causes lacunar stroke and resulting neurological syndromes. However, due to its small caliber and large individual variance, the evaluation of LSAs was limited to descriptive and objective measurements. In this study, we aimed to develop a post-processing method to quantify LSAs in subcortical regions and compare their vascular volume to conventional LSA measurements. A processing pipeline was designed to extract subcortical areas in individual spaces while screening out vessels. The vascular volume of LSAs in the subcortical region was calculated from time-of-flight-magnetic resonance angiography (TOF-MRA) at 7 Tesla. The reproducibility was tested to be good for the vascular volume (n = 5, ICCA = 0.84). Comparing the results to conventional measurements, the vascular volume was significantly correlated with the number of branches (r = 0.402, p < 0.001) and the length (r = 0.246, p = 0.032) of LSAs. By applying the method to a group of healthy volunteers (n = 40), we found that most LSAs crossing through the putamen which thereby has the highest vascular density among subcortical nuclei. In general, we proposed a semi-automated processing pipeline for quantifying the vascular volume of LSAs in subcortical regions. The novel method was tested to be robust and provided reasonable results. This method revealed spatial relationships among the perforating arteries and basal ganglia. The vascular volume can be used to evaluated blood supply of subcortical regions, benefiting the radiologic evaluation of neurodegenerative diseases caused by small vascular lesions.