OCT-tools

Segmentation and analysis of retinal layers from individual OCT b-scans, with an emphasis on computing choroid thickness. Developed by Sara Patterson (sarap44_at_uw.edu) in the Neitz lab at University of Washington.

Usage

The code works best when all the images are in a single folder with numbers as filenames (e.g., 1.png, 2.png, etc). The analysis and data will be saved with the prefix im1_, im2_, etc.

1. If you intend to compare images of the same eye, the first step is to align them. The alignImages.m function will calculate the rotation necessary to align two images and save the rotation value. Translations along the X-axis can be computed after segmentation.
2. Crop the images. Note: when the version of the image returned by the octImage property of the OCT class is rotated, then cropped so it's best to crop the image after rotating. MATLAB's built-in imcrop function is useful here.
3. Type ChoroidApp into the command line and select your image (or pass the image file path or OCT class as the first argument).
4. Segment the RPE-Choroid and ILM boundaries.
5. Use the Add Point button to add control points marking the choroid-sclera boundary. If these boundaries are not clear, open the HistogramPeakSlider UI to use 1D histograms and peak/trough detection to help with control point placement.
6. Fit the control points marking the choroid-sclera boundary with a parabola.
7. Steps 5 and 6 are usually an iterative process completed when enough control points are added to correctly fit the choroid-sclera boundary.
8. Provide a name for the exported data (works best if this includes the image ID number like im1, im2, etc). This saves .txt files of the extracted boundaries, control points and fit parameters. The OCT class properties are populated by searching for these files.
9. To compare two images, use compareChoroids.m. This function plots the two choroids and optionally computes an x-axis shift value to align the two foveal pits. The alignment is just a simple registration of maximum values. Alternatively, you can use the OCT.plotRatio() to add multiple choroid ratios to a single plot.

Dependencies

This code was written in Matlab 2018 and requires the Bioinformatics, Computer Vision, Curve Fitting, Image Processing and Signal Processing toolboxes.

References

Two external Matlab toolboxes are included: GUI Layout Toolbox and jsonlab1.5.

The initial ILM and RPE segmentation uses a simplified implementation of an algorithm introduced in:

Chiu, S.J., Li, X.T., Nicholas, P., Toth, C.A., Izatt, J.A., Farsiu, S. (2010) Automatic segmentation of seven retinal layers in SDOCT images congruent with expert manual segmentation. Optics Express, 18(18), 19413-19428

The image alignment is MATLAB's code based on:

Bay, H., Ess, A., Tuytelaars, T., Van Gool, L. (2008) SURF: Speeded up robust features. Computer Vision and Image Understanding, 110(3), 346-359