Methods and systems to detect and classify retinal structures in interferometric imaging data

A method to identify retinal structures from an optical coherence tomography (OCT) retinal image, comprising:

• (a) receiving OCT image data of a retina from an OCT system, said OCT system including a light source for generating a light beam, said light beam being divided along sample and reference paths, with the light in the sample path being directed to the retina and wherein light returning from the sample and reference paths is combined on a detector and wherein the output of the detector is supplied to a processor and wherein steps (b) to (f) are performed in a processor;
• (b) detecting edges or regions in said image data;
• (c) connecting said edges or regions into segments;
• (d) grouping two or more segments together using a segment association algorithm, whereby segments with similar statistical metrics are grouped together, said statistical metrics being selected from one or more of dispersion and moment;
• (e) assigning the grouped segments to particular retinal layers or structures;
• and
• (f) storing the assignments or a further analysis thereof in a memory or displaying the results of the assignments or a further analysis thereof on a display.