Related Experiment Video
Updated: Jun 12, 2025
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
42.5K
An Artificial Intelligence Algorithm Integrated into the Clinical Workflow Can Ensure High Quality Acute Intracranial
K Villringer1, R Sokiranski2, R Opfer3
1Center for Stroke Research Berlin, Universitätsmedizin Berlin, Berlin, Germany. kersten.villringer@charite.de.
Clinical Neuroradiology
|September 26, 2024
Summary
Artificial intelligence (AI) accurately detects intracranial hemorrhage (ICH) on CT scans, matching expert performance. This AI system integrates into clinical workflows, offering rapid diagnostics for life-threatening conditions.
Area of Science:
- Radiology
- Artificial Intelligence
- Medical Imaging
Background:
- Intracranial hemorrhage (ICH) presents a critical diagnostic challenge.
- Rapid identification and treatment are crucial for patient outcomes.
Purpose of the Study:
- To evaluate the diagnostic accuracy and efficiency of an AI algorithm for detecting ICH on cranial CT (CCT) scans.
- To assess the AI's suitability for routine clinical radiological practice.
Main Methods:
- A convolutional neural network (CNN) was trained on over 674,000 CCT slices to detect ICH.
- The AI was integrated into three German pilot centers, processing real-world data.
- AI performance was compared against two expert radiologists and inter-rater agreement.
Main Results:
- The AI achieved high diagnostic performance: sensitivity 0.90, specificity 0.96, accuracy 0.96.
- AI performance was comparable or superior to expert inter-rater agreement (0.84-0.96).
- Median processing times from acquisition to AI results ranged from 9-12 minutes.
Conclusions:
- AI demonstrates high accuracy in detecting ICH, on par with expert radiologists.
- The AI system can be seamlessly integrated into clinical workflows.
- The AI offers rapid diagnostic capabilities suitable for routine radiological practice.
