Related Experiment Video
Updated: Feb 22, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
43.6K
Detection of Intracranial Hypertension using Deep Learning.
Benjamin Quachtran1, Robert Hamilton2, Fabien Scalzo1
1Department of Computer Science and Neurology, University of California, Los Angeles (UCLA).
Proceedings of the ... IAPR International Conference on Pattern Recognition. International Conference on Pattern Recognition
|September 23, 2017
Summary
Deep learning accurately detects intracranial hypertension by analyzing brain pressure waveform morphology. This AI approach improves early detection, potentially reducing patient risks associated with delayed diagnosis.
Area of Science:
- Neurology
- Biomedical Engineering
- Artificial Intelligence
Background:
- Intracranial hypertension (high brain pressure) diagnosis is reactive, occurring after pressure elevation.
- Current monitoring methods in neurointensive care carry risks of misdetection and complications.
- Accurate, early detection of intracranial hypertension is crucial for timely intervention.
Purpose of the Study:
- To investigate Deep Learning for modeling the relationship between intracranial pressure waveform morphology and hypertension.
- To develop an accurate, predictive model for detecting intracranial hypertension using waveform data.
Main Methods:
- Utilized data from 60 patients, recording intracranial pressure over 30-minute intervals.
- Processed recordings into average normalized beats within 30-second segments.
- Labeled each beat as high (>15 mmHg) or low intracranial pressure for model training and testing.
Main Results:
- The Deep Learning model achieved 92.05± 2.25% accuracy in detecting intracranial hypertension.
- The model successfully identified patterns in waveform morphology indicative of elevated intracranial pressure.
Conclusions:
- Deep Learning offers a promising, accurate method for detecting intracranial hypertension.
- Waveform morphology analysis via AI can enhance early diagnosis and patient management.
