Predicting vasospasm risk using first presentation aneurysmal subarachnoid hemorrhage volume: A semi-automated CT image segmentation analysis using ITK-SNAP

Abstract

Purpose: Cerebral vasospasm following aneurysmal subarachnoid hemorrhage (aSAH) is a significant complication associated with poor neurological outcomes. We present a novel, semi-automated pipeline, implemented in the open-source medical imaging analysis software ITK-SNAP, to segment subarachnoid blood volume from initial CT head (CTH) scans and use this to predict future radiological vasospasm.

Methods: 42 patients were admitted between February 2020 and December 2021 to our tertiary neurosciences center, and whose initial referral CTH scan was used for this retrospective cohort study. Blood load was segmented using a semi-automated random forest classifier and active contour evolution implemented in ITK-SNAP. Clinical data were extracted from electronic healthcare records in order to fit models aimed at predicting radiological vasospasm risk.

Results: Semi-automated segmentations demonstrated excellent agreement with manual, expert-derived volumes (mean Dice coefficient = 0.92). Total normalized blood volume, extracted from CTH images at first presentation, was significantly associated with greater odds of later radiological vasospasm, increasing by approximately 7% for each additional cm3 of blood (OR = 1.069, 95% CI: 1.021-1.120; p < .005). Greater blood volume was also significantly associated with vasospasm of a higher Lindegaard ratio, of longer duration, and a greater number of discrete episodes. Total blood volume predicted radiological vasospasm with a greater accuracy as compared to the modified Fisher scale (AUC = 0.86 vs 0.70), and was of independent predictive value.

Conclusion: Semi-automated methods provide a plausible pipeline for the segmentation of blood from CT head images in aSAH, and total blood volume is a robust, extendable predictor of radiological vasospasm, outperforming the modified Fisher scale. Greater subarachnoid blood volume significantly increases the odds of subsequent vasospasm, its time course and its severity.

Fig 1. Image processing pipeline describing data collection, pre-processing, segmentation, quality control and information extraction.

(yellow font = software library; DICOM = Digital Imaging and Communications in Medicine; nii = nifti file type; FSL = FMRIB Software Library; ITK-Snap = Insight Segmentation and Registration Toolkit; RF = random forest; HU = Hounsfield Units. ‘CT-scan’ designed using resources from Flatiron.com).

Fig 2. Examples of semi-automated subarachnoid hemorrhage segmentations for each modified Fisher grade.

(yellow = segmentation overlay, mFS = modified Fisher Scale).

Fig 3. Greater segmented blood load is associated with greater radiological vasospasm risk.

Boxplot of blood volume in patients who developed radiological vasospasm (maroon) and those who did not (blue). (*** = p< .001).

Fig 4. Receiver operating characteristic (ROC) curve demonstrating the performance characteristics of the binary classifier fit in the logistic regression model.

Black = univariate logistic regression model using normalized total blood volume; gray = logistic regression model using dummy-coded modified Fisher score values.

Fig 5. Associations between subarachnoid blood volume and metrics of severity of radiological vasospasm.

A: Boxplots of blood volume grouped by radiologist’s impression of subjective severity of vasospasm. B: Scatter plot showing the greatest recorded Lindegaard ratio from TCD plotted against normalized blood volume. (* = p < .05, ** = p < .01).

Fig 6. Associations between subarachnoid blood volume and temporal vasospasm-related outcomes.

A: Scatter plot showing the number of discrete episodes of vasospasm against normalized blood volume. B: Scatter plot showing the duration of vasospasm in days plotted against normalized blood volume. C: Scatter plot showing the total length of hospital stay plotted against normalized blood volume.