Developing polarization sensitive micro optical coherence tomography – towards high contrast, resolution and sensitivity optical imaging

• Main Author: Liu, Xinyu
• Other Authors: Liu Linbo
• Format: Theses and Dissertations
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/73181
• Institution: Nanyang Technological University

Optical coherence tomography (OCT) has become established as a high-resolution three-dimensional imaging modality for the diagnosis of various human diseases. This thesis focuses on the technical development of current OCT systems, and proposes several improvements to OCT in terms of contrast, resolution and sensitivity. Polarization sensitive-OCT (PS-OCT) needs two different polarization states illuminating on the same sample spot to solve the local birefringence properties, specifically, local retardation and local relative optic axis. Current PS-OCT uses a polarization modulator to change the polarization state of adjacent A-scan lines, or encodes various polarization detections on different imaging depth. However, these methods are not applicable for high resolution OCT. The bandwidth of a modulator limits the system resolution lower than about 5 µm. The depth range of an OCT is determined by the wavelength resolution of the spectral detection. For high resolution OCT with an ultra-broadband source, there are little margins left for extra depth multiplexing. To overcome these problems, this thesis describes a new reasoning methodology, derived from the polarization reciprocity theorem, termed as “holding point reasoning”. This reasoning theorem states that, for a sample with random birefringence layer structures (no diattenuation), the backscattered polarization state evolving trace along the depth is consisted discontinuous arcs lying on the Poincare sphere. In a unitary backscattering system, the circles that the arcs are on all pass through the “holding point”, which is the Q-U plane reflection of the input polarization state. In the second chapter of this thesis, I described both the experiments and theoretical confirmation of this reasoning method. This reasoning technique enables the single input PS-OCT to resolve local birefringence properties, open a way towards the construction of high resolution polarization sensitive optical coherence tomography. In this chapter, the prototype of such system based on holding point reasoning was demonstrated. Collaborating with one of the leading research groups, I developed an approach to fabricate a tissue-like OCT birefringence phantom, a key protocol for characterizing and calibrating the absolute detection value, sensitivity and range of PS-OCT. Taking use of the stress-induced birefringence, the leveled birefringence regions were created by stretching the polycarbonate sheet to different lengths. Several pieces of polycarbonate sheet with different birefringence and different patterns were fabricated. To package the target, the birefringent samples were embedded into matrix made of epoxy. Scatters were added into the polycarbonate and epoxy to mimic the scattering property of the biological tissue. The fabrication, test and demarcation of the target were described in the third chapter. In fiber-based OCT, the interference fringes suffer from the fading effect due to misalignment of the light polarization states between the reference and sample arms, resulting in sensitivity degradation and image intensity variation. We theoretically and experimentally analyzed the relation between the misalignment and the fading coefficient. Assuming that the variation of the light polarization in single mode fiber was a random process, we statistically quantified the fading effect. Furthermore, in OCT configuration based on Michelson interferometer, we reported an interesting observation, that the polarization states of light travelling a round trip in SMF are not evenly distributed on the Poincare sphere. Based on this observation, we demonstrated the existence of an optimal output polarization state of the reference arm to mitigate the fading effect. We demonstrated that in an optimal setup, the statistical average SNR could be 3.5 dB higher than a setup without proper polarization management. The fourth chapter of the thesis discusses the management of polarization fading problem in OCT. Besides from polarization detection, to maintain and improve the resolution of OCT, I developed a spectral estimation (SE-OCT) algorithm to super-resolve the layer structures in transparent samples. This algorithm extrapolates the interferometric fringes outside the source bandwidth, thereby improving the resolving ability of OCT. Based on the assumption that the sample was consisted of sparse layers, the algorithm can super resolve two close surfaces within 200 nm, far under the physical resolution. SE-OCT shows good performance in transparent tissues such as corneal. The discussion of SE-OCT is included in the fifth chapter. In this thesis, a method to improve the sensitivity of spectral domain OCT is also included. Termed as spectral encoded extended source OCT (SEES-OCT), this new configuration is an attempt to improve signal strength for ophthalmic imaging. By introducing a dispersive element in the infinity space of the sample arm, the scanning laser spot was dispersed with a visual angle of 7.9 mrad. The maximum permissible exposure (MPE) of such an extended source is 3.1 times larger than that of a “standard” point source OCT, which corresponds to sensitivity improvement of 5 dB. The advantage of SEES-OCT in providing superior penetration depth over a point source system was demonstrated. In conclusion, this thesis provides a step forward to develop a viable OCT with high resolution, sensitivity and contrast with the discussion and solutions to several technical problems. The feasibility and effectiveness of the proposed solutions were demonstrated with phantoms and ex vivo animal imaging in the context of relevant biological diseases.