A validated web-application (GFDC) for automatic classification of glaucomatous visual field defects using...

Area of Science:

• Ophthalmology
• Medical Technology
• Computational Medicine

Background:

• Glaucoma diagnosis relies on visual field classification, which is often subjective and time-consuming.
• Current standardized methods for classifying glaucomatous visual field defects are labor-intensive, limiting their practical clinical use.
• The accuracy and reliability of glaucoma diagnosis, prognostication, and management are hindered by the subjectivity of visual field assessments.

Purpose of the Study:

• To introduce and validate a novel web-application, the Glaucoma Field Defect Classifier (GFDC).
• To enable the automatic application of the Hodapp-Parrish-Anderson classification system for glaucomatous visual field defects.
• To improve the accuracy and fairness of clinical decision-making in glaucoma patient care.

Main Methods:

• Development of a web-application named Glaucoma Field Defect Classifier (GFDC).
• Validation of the GFDC using a cross-sectional study design.
• Implementation of automatic classification rules based on the Hodapp-Parrish-Anderson criteria.

Main Results:

• The Glaucoma Field Defect Classifier (GFDC) demonstrated perfect accuracy in classifying visual field defects.
• GFDC accurately distinguished between mild, moderate, and severe glaucomatous field defects.
• The application provides a reliable and automated method for visual field defect classification.

Conclusions:

• The Glaucoma Field Defect Classifier (GFDC) offers a highly accurate and efficient solution for classifying glaucomatous visual field defects.
• GFDC has the potential to significantly improve the accuracy and consistency of clinical decisions in glaucoma management.
• The freely available online application and its source code empower clinicians and researchers in glaucoma patient care.