Background: There is no facile quantitative method for monitoring hydrocephalus (HCP). We propose quantitative computed tomography (CT) ventriculography (qCTV) as a novel computer vision tool for empirically assessing HCP in patients with subarachnoid hemorrhage (SAH).
Methods: Twenty patients with SAH who were evaluated for ventriculoperitoneal shunt (VPS) placement were selected for inclusion. Ten patients with normal head computed tomography (CTH) findings were analyzed as negative controls. CTH scans were segmented both manually and automatically (by qCTV) to generate measures of ventricular volume.
Results: The median manually calculated ventricular volume was 36.1 cm3 (interquartile range [IQR], 30-115 cm3), which was similar to the median qCTV measured volume of 37.5 cm3 (IQR, 32-118 cm3) (P = 0.796). Patients undergoing VPS placement demonstrated an increase in median ventricular volume on qCTV from 21 cm3 to 40 cm3 on day T-2 and to 51 cm3 by day 0, a change of 144%. This is in contrast to patients who did not require shunting, in whom median ventricular volume decreased from 16 cm3 to 14 cm3 on day T-2 and to 13 cm3 by day 0, with an average overall volume decrease 19% (P = 0.001). The average change in ventricular volume predicted which patients would require VPS placement, successfully identifying 7 of 10 patients (P = 0.004). Using an optimized cutoff of a change in ventricular volume of 2.5 cm3 identified all patients who went on to require VPS placement (10 of 10; P = 0.011).
Conclusions: qCTV is a reliable means of quantifying ventricular volume and hydrocephalus. This technique offers a new tool for monitoring neurosurgical patients for hydrocephalus, and may be beneficial for use in future research studies, as well as in the routine care of patients with hydrocephalus.
Keywords: Computer vision; Hydrocephalus; Imaging; Informatics; Subarachnoid hemorrhage.
Copyright © 2017 Elsevier Inc. All rights reserved.