Objective   Spectral domain optical coherence tomography (SD-OCT) is a technique used for tomography and three-dimensional imaging of the internal microstructure of materials and biological tissues. By performing Fourier transform on the collected interference signal, the SD-OCT system can reconstruct image depth information, providing a detailed view of the sample's structure. However, the resulting image includes a mirror image with zero phase delay symmetry, which can be problematic. To address this issue, the sample is positioned on the zero phase delay side during scanning to eliminate the influence of the mirror image. This approach, while effective, only utilizes half of the imaging range space. Improving the detection depth of OCT systems by eliminating complex conjugate images is crucial for enhancing the accuracy and reliability of the imaging process.

  Methods   The deconjugated imaging system was developed from the original SD-OCT system (Fig.3). The system utilizes a broadband light source, with the light being split by a beam splitting cube into the reference arm and sample arm. The reference arm consists of a collimator and a reflector, while the sample arm includes a two-dimensional galvanometer, scanning lens, and sample stage. The two-dimensional galvanometer enables scanning of the sample in both X and Y directions by deflecting the reflector. The entire sample arm system is finely tuned using an electronic displacement platform to adjust the relative position, create beam offset, and add carrier frequency. The sample stage can be moved axially to focus imaging at different depths within the sample, allowing for the superposition of focusing planes and enhancing the imaging depth.

  Results and Discussions   To eliminate the influence of the complex conjugate image, the distance between the beam and the center of the mirror is adjusted to 1.2 mm. Before the spectral data undergoes the standard image reconstruction process, the complex conjugate image is removed using the added carrier frequency through Hilbert transform (Fig.5). This results in a doubling of the imaging range compared to the original image. However, the focal depth of the lens is much smaller than the imaging depth, leading to a decrease in lateral resolution outside the depth of field (DOF) range. By focusing on imaging samples at different depths, extracting the focal plane, and using Gaussian weighted fusion (Fig.6), the imaging depth of the system is doubled from 1.97 mm to 3.94 mm, without being limited by the lens depth of focus. The sample structure information is clearly visible (Fig.7). When applied to biological tissues, the beam penetration depth is limited by the scattering and absorption properties of the tissues. The imaging depth is increased from 0.51 mm to 0.65 mm, with well-connected tissue signals in the vertical direction. The strip-like structure of the colon wall recess is clearly visible (Fig.8).

  Conclusions   A method has been developed to improve the depth and quality of OCT imaging by combining the carrier frequency method to remove conjugate images and beam focusing at different depths of the sample. This method was tested on various samples, including multi-layer transparent films and biological tissues. Results showed that the effective imaging depth for transparent films doubled to 3.94 mm, while for colon tissue, the imaging depth increased from 0.51 mm to 0.65 mm, a 27.5% increase. The extended depth also allowed for clearer display of deep structural information in the sample.