Related Experiment Video
Updated: May 14, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Non-invasive cerebrospinal fluid pressure estimation using multi-layer perceptron neural networks.
S Mojtaba Golzan1, Alberto Avolio, Stuart L Graham
1Australian School of Advanced Medicine, Macquarie University, Australia. mojtaba.golzan@ieee.org
Summary
This study introduces a non-invasive method to estimate cerebrospinal fluid pressure (CSFp) using retinal venous pulsation and intraocular pressure (IOP). This technique offers a safer alternative to invasive CSFp measurements, potentially improving patient outcomes.
Area of Science:
- Ophthalmology
- Neurology
- Biomedical Engineering
Background:
- Cerebrospinal fluid pressure (CSFp) is crucial for diagnosing neurological conditions like hydrocephalus and brain tumors.
- Current invasive CSFp measurement methods carry risks of infection and morbidity.
- A physiological link exists between CSFp, intraocular pressure (IOP), and retinal venous pulsation amplitude.
Purpose of the Study:
- To develop and validate a non-invasive method for estimating CSFp.
- To leverage the relationship between retinal venous pulsation, IOP, and CSFp.
- To create an algorithm for estimating CSFp using artificial neural networks.
Main Methods:
- Recruited 15 subjects for the study.
- Measured dynamic retinal venous diameter changes and IOP.
- Utilized artificial neural networks (ANN) to model the relationship between inputs (retinal venous pulsation amplitude, IOP) and output (CSFp).
Main Results:
- ANN model achieved a mean square error of 2.4 mmHg for training data and 1.27 mmHg for testing data.
- No significant difference was found between experimentally measured and ANN-estimated CSFp values (p>0.01).
- Demonstrated the feasibility of using retinal venous pulsatility and IOP to estimate CSFp.
Conclusions:
- Non-invasive estimation of CSFp is achievable using retinal venous pulsatility and IOP.
- This novel approach may reduce risks associated with invasive CSFp monitoring.
- Further research could validate this method for clinical application in neurological disorder management.
