Analyzing OCT-based methods of depth-resolved optical attenuation mapping using digital phantoms

Abstract

One of the main depth-resolved parameters extracted from structural scans in optical coherence tomography (OCT) is the optical attenuation coefficient (OAC). However, for OAC mapping, researchers utilize various methods, each with its own strengths and weaknesses. Creating physical phantoms with sufficiently controllable properties (controllable values of the attenuation coefficient and levels of additive and speckle noises) for analysing OAC methods is a rather challenging problem. Therefore, a comparison of the main methods used for OAC mapping presented in this study is made using realistically-simulated OCT scans with known specified parameters. Such digital phantoms have a number of advantages: (i) there is no influence of the device's hardware function (e.g. the signal decay due to influence of the distance from the receiving system, including the influence of OCT-beam focusing and the roll-off effect for spectral-domain OCT devices), (ii) it has the ability to vary the attenuation coefficient to nearly perfectly match real scans, (iii) arbitrary scenarios of spatial inhomogeneity of OAC can readily be realized. This allows one to investigate OAC mapping methods for a wide range of possible OCT images. The comparison of the methods was carried out using the same size of sliding averaging windows, the use of which is very important due to the presence of speckle noise in real OCT scans. The analysis can be useful in the development of improved OCT-based methods for mapping the attenuation coefficient. The study was supported by RSF grant No 23-72-01107.

Speaker

Peter A. Chizhov
A.V. Gaponov-Grekhov Institute of Applied Physics of the Russian Academy of Sciences
Russia

Discussion

Ask question