Related Experiment Videos
[Development of an automatic discrimination system for glaucomatous visual fields based on neuro-fuzzy nets]
J García Feijoó1, E J Carmona Suárez, L M Gallardo
1Hospital Clínico San Carlos, Instituto de Investigaciones Oftalmológicas Ramón Castroviejo, Universidad Complutense de Madrid. España. mherrerad@sego.es
Archivos De La Sociedad Espanola De Oftalmologia
|December 10, 2002
Summary
This study developed an automatic system using neuro-fuzzy rules to classify visual fields for glaucoma diagnosis. The system achieved high accuracy, offering a valuable tool for detecting glaucoma damage.
Area of Science:
- Ophthalmology
- Artificial Intelligence
- Medical Diagnostics
Background:
- Glaucoma diagnosis relies on visual field examination.
- Automated classification systems can aid in early detection and management.
- Neuro-fuzzy logic offers a framework for complex rule-based systems.
Purpose of the Study:
- To develop an automatic visual field classification system for glaucoma diagnosis.
- To utilize neuro-fuzzy rules for enhanced diagnostic accuracy.
- To create a tool for distinguishing between normal and glaucomatous visual fields.
Main Methods:
- Analysis of 212 visual fields from 198 patients (61 normal, 151 glaucomatous).
- Development of a neuro-fuzzy classifier (NEFCLASS) using mean visual field defects.
- Implementation of a learning process to optimize classifier parameters.
Main Results:
- The neuro-fuzzy classifier achieved a sensitivity of 96.0% and specificity of 93.4%.
- Five neuro-fuzzy rules were generated for classification.
- The system effectively classified visual fields based on defined input units.
Conclusions:
- Neuro-fuzzy rules provide a robust method for visual field classification in glaucoma.
- The developed system's performance is comparable to established techniques like discriminant analysis and neural networks.
- This approach offers a valuable tool for glaucoma diagnosis and management.