Validation of Automated Software for Guttae Quantification in Fuchs Endothelial Corneal Dystrophy.

Area of Science:

• Ophthalmology
• Medical Imaging
• Computational Pathology

Background:

• Corneal guttae are characteristic of Fuchs endothelial corneal dystrophy (FECD).
• Accurate quantification tools are needed for advancing FECD therapies.
• Automated analysis of guttae and their progression is crucial for patient management.

Purpose of the Study:

• To develop and validate a novel automated software platform for quantifying the guttae area ratio (GAR) in FECD specular microscopy images.
• To compare automated GAR measurements with expert manual grading.
• To assess the correlation between automated GAR and FECD disease severity and subclinical edema.

Main Methods:

• Prospective enrollment of FECD patients (mKg 2-5) without prior keratoplasty.
• Specular microscopy imaging and Scheimpflug imaging for subclinical edema assessment.
• Validation study comparing automated GAR with manual expert grading and correlating GAR with disease grade.

Main Results:

• Automated GAR demonstrated equivalence to manual expert grading (P = 0.003).
• Automated GAR significantly increased with worsening FECD disease severity (mKg 2-5).
• High repeatability of GAR measurements was observed (intraclass correlation = 0.85).

Conclusions:

• Automated software for guttae quantification in FECD is validated and equivalent to manual grading.
• Automated GAR is an objective metric that correlates with disease severity and may precede subclinical edema.
• This software enables precise, objective measurement of FECD progression and therapeutic efficacy.