Towards retrieving dispersion profiles using quantum-mimic Optical Coherence Tomography and Machine Learnin

TL;DR

This paper introduces a machine learning approach to extract dispersion profiles in quantum-mimic Optical Coherence Tomography by analyzing artefacts caused by autocorrelation, enabling non-invasive depth profiling of layered materials.

Contribution

It demonstrates a novel method that uses neural networks to determine dispersion profiles from artefacts in OCT images, linking artefact characteristics to layer dispersion properties.

Findings

Neural network accurately predicts GVD in simulated data.

Method successfully applied to experimental glass samples.

Artefact shape correlates with layer dispersion characteristics.

Abstract

Artefacts in quantum-mimic Optical Coherence Tomography are considered detrimental because they scramble the images even for the simplest objects. They are a side effect of autocorrelation which is used in the quantum entanglement mimicking algorithm behind this method. Interestingly, the autocorrelation imprints certain characteristics onto an artefact - it makes its shape and characteristics depend on the amount of dispersion exhibited by the layer that artefact corresponds to. This unique relationship between the artefact and the layer's dispersion can be used to determine Group Velocity Dispersion (GVD) values of object layers and, based on them, build a dispersion-contrasted depth profile. The retrieval of GVD profiles is achieved via Machine Learning. During training, a neural network learns the relationship between GVD and the artefacts' shape and characteristics, and consequently, it is able to provide a good qualitative representation of object's dispersion profile for never-seen-before data: computer-generated single dispersive layers and experimental pieces of glass.