Basis function model to extract the combined confocal and fall-off function from multiple optical coherence tomography A-scans.

Significance: Many derivatives of optical coherence tomography (OCT) rely on the depth-dependent information of the sample in the image. System depth-dependent effects, such as the confocal effect and the sensitivity fall-off, should be corrected to improve the accuracy of the images and information derived from them. Aim: We developed a new single-shot method to extract the combined confocal and fall-off functions and remove system-generated depth-dependent effects from OCT images. Approach: The combined function is modeled as a linear combination of basis functions whose coefficients are found from two or more A-scans (or B-scans) of a sample that are vertically shifted within the imaging range. No prior knowledge of the OCT system parameters or assumed form for the confocal and fall-off functions is needed. Results: The method was derived and validated with simulations and OCT images of a phantom, a biological sample, and human retina. Improvement over the Ratio Fit method was demonstrated. Conclusions: The improvement in the extraction of the combined confocal and fall-off effects by this method should lead to improved medical diagnosis through more accurate attenuation coefficient calculations. The method enables future applications of OCT where precise removal of all depth-dependent effects on OCT images is critical.