Establishing a reliable visual function test and applying it to screening optic nerve disease in onchocercal...

Area of Science:

• Ophthalmology
• Computer Science
• Biomedical Engineering

Background:

• Computer Controlled Video Perimetry (CCVP) is vital for detecting visual function loss from conditions like onchocerciasis and glaucoma.
• Portable CCVP systems show high acceptability in field studies but struggle with reliability due to behavioral variants and inconsistent environments.
• Existing CCVP methods face challenges in delivering accurate results in non-standardized testing conditions.

Purpose of the Study:

• To propose and validate a novel architecture for a more reliable computer-controlled video perimetry (CCVP) system.
• To address the issue of measurement noise and unreliability in portable CCVP testing.
• To enhance the accuracy and consistency of visual field testing in challenging field settings.

Main Methods:

• Implementation of a self-organising neural network to mitigate measurement noise from human behavioral factors.
• Integration of a dedicated control unit to manage the overall system behavior and testing process.
• Deployment of the integrated system for screening optic nerve disease in onchocercal communities in rural Nigeria.

Main Results:

• The proposed CCVP architecture demonstrated encouraging experimental results across a large number of test records.
• The system successfully obtained reliable visual field test results despite volatile testing environments.
• The application of self-organising neural networks effectively managed behavioral-induced measurement noise.

Conclusions:

• The developed CCVP system architecture offers a reliable solution for visual function loss detection in field settings.
• The proposed method enables the acquisition of dependable results from portable perimetry systems, even in non-ideal conditions.
• This approach holds significant promise for improving the diagnosis and management of optic nerve diseases in resource-limited areas.