Related Experiment Video
Updated: Jul 18, 2026
07:51
Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts
Published on: October 21, 2022
Endothelial cell density in donor corneas: a comparison of automatic software programs with manual counting
Christoph Hirneiss1, Ricarda G Schumann, Martin Grüterich
1Department of Ophthalmology, Ludwig-Maximilians-Universität, Munich, Germany.
Cornea
|January 2, 2007
Summary
Automated systems for endothelial cell density (ECD) analysis show varied accuracy. The NAVIS system’s fully automated mode demonstrated significant deviations from manual counts, highlighting the need for validation.
Area of Science:
- Ophthalmology
- Biomedical Engineering
- Cell Biology
Background:
- Automated systems are increasingly used for endothelial cell density (ECD) assessment in donor corneas.
- Accurate ECD measurement is crucial for corneal transplantation success.
Purpose of the Study:
- To compare the accuracy of two automated systems, Endothelial Analysis System EAT V1.4 and NAVIS Cell Count V3.4.1, against manual ECD counts.
- To evaluate the performance of semiautomatic and fully automatic analysis modes.
Main Methods:
- Manual ECD counts were performed by three ophthalmologists on 50 human corneas.
- Digital images were analyzed using the semiautomatic EAT system V1.4.
- Images were analyzed using the fully automatic NAVIS system V3.4.1.
- Direct comparison of NAVIS counts with manual counts was conducted on 11 additional corneas.
Main Results:
- The EAT system showed a non-significant 1.3% increase in ECD compared to manual counts.
- The fully automatic NAVIS system showed a significant 7.0% increase in ECD (P = 0.0001).
- Altman-Bland analysis indicated that the fully automatic NAVIS system produced inconsistent results across different ECD ranges.
Conclusions:
- Fully automated analysis of endothelial cell density presents challenges and requires careful validation.
- Computer-assisted ECD measurement systems must be validated against established manual counting methods.
