Related Experiment Video
Updated: Mar 30, 2026
11:29
Corneal Confocal Microscopy: A Novel Non-invasive Technique to Quantify Small Fibre Pathology in Peripheral Neuropathies
Published on: January 3, 2011
27.5K
Reproducibility of In Vivo Corneal Confocal Microscopy Using an Automated Analysis Program for Detection of Diabetic
Ilia Ostrovski1, Leif E Lovblom1, Mohammed A Farooqi1
1Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada.
Plos One
|November 6, 2015
Summary
A new automated analysis program for in vivo corneal confocal microscopy (IVCCM) shows high reproducibility for detecting diabetic sensorimotor polyneuropathy (DSP). This automated method offers a promising alternative to manual analysis for corneal nerve fibre length (CNFL) measurements.
Area of Science:
- Ophthalmology
- Neurology
- Diabetology
Background:
- In vivo corneal confocal microscopy (IVCCM) is a validated, non-invasive method for detecting diabetic sensorimotor polyneuropathy (DSP).
- Current IVCCM image analysis is time-consuming and requires specialized expertise, limiting its widespread clinical utility.
- Automated analysis programs aim to improve the efficiency and reproducibility of IVCCM image interpretation.
Purpose of the Study:
- To evaluate the inter- and intra-observer reproducibility of a novel automated analysis program for IVCCM.
- To compare the automated analysis results with manual analysis of corneal nerve fibre length (CNFL).
- To assess the utility of automated IVCCM for DSP detection.
Main Methods:
- A cross-sectional diagnostic study involving 20 non-diabetes controls and 26 participants with type 1 diabetes.
- Participants underwent two separate IVCCM examinations by one observer and a third by an independent observer.
- Corneal nerve fibre length (CNFL) was measured using manual analysis (CNFLMANUAL), semi-automated analysis (CNFLSEMI-AUTOMATED), and fully-automated analysis (CNFLFULLY-AUTOMATED). Reproducibility was assessed using intraclass correlation coefficients (ICC).
Main Results:
- The fully-automated analysis (CNFLFULLY-AUTOMATED) demonstrated the highest intra-observer reproducibility (ICC=0.84) compared to semi-automated (ICC=0.74) and manual (ICC=0.72) methods.
- Inter-observer reproducibility was comparable across all methods, with ICCs ranging from 0.73 to 0.78.
- Automated methods (semi- and fully-automated) underestimated CNFL compared to manual analysis, but fully-automated analysis showed better preservation of reproducibility.
Conclusions:
- Fully-automated IVCCM analysis of CNFL maintains high reproducibility, despite a measurement bias compared to manual analysis.
- This automated approach offers a more efficient and reproducible method for assessing corneal nerves in diabetic neuropathy.
- Further research is needed to establish diagnostic thresholds for the fully-automated CNFL measure in DSP detection.
