Optical coherence tomography depth reconstruction method based on deep learning

Abstract

The invention discloses an optical coherence tomography depth reconstruction method based on deep learning. The method comprises the following steps: 1) preparing a training sample which comprises interference signal data and corresponding real depth data; 2) training a neural network to learn a mapping relationship between an interference signal and a real depth; 3) obtaining an interference signal of a to-be-reconstructed depth and performing depth reconstruction by using the model. According to the invention, deep learning is used to realize deep reconstruction (not a Fourier transform method) of an interference signal acquired in an optical coherence tomography process; the depth reconstruction resolution can be not limited by the central wavelength or spectral width of a light source,no special requirement is needed for the smoothness of a spectral envelope, and an interference signal can be obtained through a simplified SD-OCT method of removing a reference arm.

Description

technical field [0001] The invention belongs to the field of computational optical imaging, in particular to a deep learning-based optical coherence tomography depth reconstruction method. Background technique [0002] Spectral-Domain Optical Coherence Tomography (SD-OCT for short) is an optical three-dimensional imaging technology that uses a broadband light source. The broadband optical signal is scattered back from the surface and inside of the sample and interferes. The intensity and phase information of the scattered light is encoded in the interference signal. The depth reconstruction process is to record each spectral component of the interference signal separately, and then reconstruct the depth-resolved interference envelope signal through Fourier transform. Through the Fourier transform of the interference spectrum, the one-dimensional depth of the sample can be obtained. Information, combined with lateral scanning, can realize three-dimensional imaging of the sam...

AI Technical Summary

Problems solved by technology

This limits the choice of light sources available, or adds complex steps to smoothing the spectral signal

[0006] (3) Ordinary OCT uses a reference arm to suppress the spurious signal strength, and the reference arm will increase the complexity of the device and make the measurement results susceptible to vibration interference

The reference arm requires the same optical path as the sample

Images