Related Experiment Video
Updated: May 13, 2025
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
42.4K
Estimation of Ventricular and Intracranial Hemorrhage Volumes and Midline Shift on an External Validation Data Set
Marco Colasurdo1, Dor Amran2, Huanwen Chen3
1Department of Interventional Radiology, Oregon Health and Science University, Portland , Oregon , USA.
Neurosurgery
|April 14, 2025
Summary
An AI algorithm, Viz ICH-Plus, accurately quantifies intracranial hemorrhage (ICH), bilateral lateral ventricular (BLV) volumes, and midline shift (MLS) on CT scans, aiding clinical decision-making.
Area of Science:
- Neuroradiology
- Medical Imaging Analysis
- Artificial Intelligence in Medicine
Background:
- Noncontrast head CT is crucial for managing intracranial hemorrhage (ICH).
- Manual volume and shift measurements are time-consuming.
- Automated quantification tools are needed to improve efficiency.
Purpose of the Study:
- To evaluate the performance of the Viz ICH-Plus AI algorithm.
- To assess its accuracy in quantifying ICH, BLV volumes, and MLS.
- To determine its ability to detect clinically significant changes in ICH management.
Main Methods:
- Retrospective analysis of 78 head CT scans from 30 ICH patients.
- Comparison of Viz ICH-Plus automated measurements with manual contours.
- Assessment of median absolute errors and detection of clinically meaningful changes.
Main Results:
- Viz ICH-Plus demonstrated low median absolute errors for ICH (2.9 mL), BLV (5.3 mL), and MLS (1.1 mm).
- The algorithm showed high accuracy (92.3%) in detecting clinically significant changes between scans.
- Sensitivity and specificity for change detection were 91.7% and 92.6%, respectively.
Conclusions:
- Viz ICH-Plus offers a reliable AI solution for quantifying ICH, BLV volumes, and MLS.
- The algorithm shows good performance and predictive power for clinically significant changes.
- This tool can potentially streamline ICH management workflows.
