Low-cost and digital refocusing approaches to high-resolution spectral-domain optical coherence tomography

• Main Author: Wang, Lulu
• Other Authors: Liu Linbo
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Electrical and electronic engineering::Optics, optoelectronics, photonics
• Online Access: https://hdl.handle.net/10356/144267
• Institution: Nanyang Technological University

Imaging pathophysiological processes noninvasively at the cellular and subcellular level are critical for understanding and diagnosis of human diseases. High-resolution optical coherence tomography (OCT), including but not limited to ultrahigh-resolution OCT (UHR OCT) and micro-OCT (µOCT), are particularly suitable for this purpose owning to their superb spatial resolutions. However, technical difficulties have to be overcome for the translation of these technologies for clinical use. First of all, the penetration depth of the current high-resolution OCT, co-determined by the focal depth, ranging depth, and the sensitivity, is insufficient. Secondly, there is a compelling need to lower the cost of devices so that the tests are affordable to more people. The sensitivity limitation of the current µOCT is partly due to the low transmission efficiency of the refractive spectrometer optics. I developed a low-cost, high transmission efficiency all-reflective-optics-based spectrometer, which provides 27% better transmission efficiency without any compromise in the ranging depth and the axial resolution. In addition, the cost of the camera lens is reduced by 2/3. The cost of the spectrometer is reduced by 50% with respect to a commercial spectrometer with similar performances. To extend ranging depth, I developed a dual-spectrometer system. To combine the spectra acquired at the two line scan cameras, I developed a fully automatic algorithm to search for the overlapping region in the spectral interferograms. A total ranging depth of 3.5 mm and a 6-dB roll-off range of 1.44 mm were demonstrated experimentally, which is 94% and 78% larger than a single spectrometer system, respectively. In combination with a forward-model based digital refocusing method, I demonstrate that this system is capable of visualizing the full thickness of the pig cornea over the depth of 2.2 mm (n = 1.375) and visualize corneal endothelial cells over a depth range of 120 µm ex vivo. To mitigate the focal depth requirement, I developed a new digital refocusing method, where I use digital phase pupil filters to improve the transverse resolution. The method is based on the multiple aperture synthesis OCT setup, and digital phase masks were selected using the guide star approach. We experimentally observed 24% improvement in the mainlobe width, and the sidelobe problem is suppressed by 1.1 dB through averaging results from multiple distinct phase masks. In conclusion, I have developed and validated three optical imaging methods to address the penetration limitation of high-resolution OCT. My first two efforts in improving the spectrometer design mainly restore the signal strength by reducing the transmission loss and the path length-dependent signal “roll-off”, respectively. These two methods can be combined to an optimal dual-spectrometer design, where a reflective parabolic collimator and low-cost refractive camera lens are used in each spectrometer. My efforts in digital refocusing OCT can be potentially used to mitigate the focal range limitation of high-resolution OCT. My future work will be focused on the development of an advanced µOCT system featured with transverse super-resolution and high-efficiency, low-cost spectrometers.