Related Experiment Video
Updated: Mar 14, 2026
07:44
In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography
Published on: July 24, 2020
3.6K
Atlas-based shape analysis and classification of retinal optical coherence tomography images using the functional
Sieun Lee1, Nicolas Charon2, Benjamin Charlier3
1School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, V5A 1S6, Canada.
Medical Image Analysis
|October 1, 2016
Summary
We developed a new method, functional shape (fshape) analysis, to measure shape changes in retinal optical coherence tomography images. This approach helps visualize and quantify retinal nerve fiber layer loss in glaucoma patients.
Area of Science:
- Ophthalmology
- Medical Imaging
- Biomedical Engineering
Background:
- Quantitative shape variability analysis is crucial for understanding diseases like glaucoma.
- Optical coherence tomography (OCT) provides detailed retinal imaging but requires robust analysis methods.
- Existing methods may not fully capture the complex geometry and functional aspects of retinal layers.
Purpose of the Study:
- To introduce and validate the functional shape (fshape) framework for quantitative shape variability analysis in retinal OCT images.
- To develop a method for registering anatomical shapes using surface geometry and functional measures.
- To demonstrate the clinical application of fshape in analyzing the Retinal Nerve Fiber Layer (RNFL) in glaucoma.
Main Methods:
- The fshape framework was employed, integrating surface geometry with functional measures (e.g., retinal layer thickness).
- A population mean template of geometry-function measures was generated for registration.
- Shape variability was quantified using geometrical deformation and functional residual.
- Atlases of RNFL inner layer surface and thickness were created for normal and glaucomatous subjects.
- Regularized linear discriminant analysis was used for automated classification of glaucoma, glaucoma-suspect, and control cases.
Main Results:
- The fshape framework successfully generated population mean templates and quantified shape variability.
- Atlases revealed detailed spatial patterns of RNFL loss in glaucoma.
- The classification model achieved automated differentiation between glaucoma, glaucoma-suspect, and control groups based on RNFL fshape metrics.
Conclusions:
- The fshape framework offers a novel and effective approach for quantitative shape variability analysis in retinal OCT.
- This method provides valuable insights into the spatial patterns of RNFL loss in glaucoma.
- Fshape metrics show potential for automated diagnosis and monitoring of glaucoma.
