Related Experiment Video
Updated: Apr 26, 2026
Endovascular Perforation Model for Subarachnoid Hemorrhage Combined with Magnetic Resonance Imaging MRI
Published on: December 16, 2021
Automatic quantification of subarachnoid hemorrhage on noncontrast CT
A M Boers1, I A Zijlstra2, C S Gathier3
1From the Departments of Radiology (A.M.B., I.A.Z., R.v.d.B., H.A.M., C.B.M.) Biomedical Engineering and Physics (A.M.B., H.A.M.), Academic Medical Center, Amsterdam, the Netherlands Institute of Technical Medicine (A.M.B.) a.m.boers@amc.uva.nl.
This study developed an accurate, automated method to quantify subarachnoid hemorrhage (SAH) volume and density from non-contrast CT scans. The new technique shows high correlation with manual measurements, offering potential for improved patient outcome prediction.
Area of Science:
- Neuroradiology
- Medical Imaging Analysis
- Computational Pathology
Background:
- Quantification of subarachnoid hemorrhage (SAH) on non-contrast computed tomography (NCCT) is crucial for predicting patient outcomes and guiding treatment.
- Current methods do not account for hemorrhage volume and density variations.
- There is a need for automated, accurate quantification of SAH parameters.
Purpose of the Study:
- To develop and validate a fully automatic method for quantifying SAH volume and density.
- To improve upon existing radiologic measures for SAH assessment.
- To provide a tool for more precise SAH analysis in clinical practice and research.
Main Methods:
- An automatic method was developed based on relative density increase in NCCT, accounting for partial volume effects and beam-hardening.
- The method was validated by comparing automatic volume and density measurements with manual delineations from 30 patients.
- Intraclass correlation coefficient and Bland-Altman analysis were used to assess agreement with manual measurements and interobserver variability.
Main Results:
- The automatic method successfully segmented SAH in all 30 patients.
- High correlation was observed between automatic and manual measurements for both SAH volume (ICC=0.98) and density (ICC=0.80).
- The automatic method demonstrated excellent accuracy with narrow limits of agreement, comparable to interobserver variability.
Conclusions:
- The developed automatic method for SAH volume and density quantification is highly accurate compared to manual assessment.
- This automated approach has the potential to provide significant determinants for clinical decision-making and research in SAH.
- The method offers a reliable and efficient alternative for SAH analysis in neuroimaging.
