Design of flexible multi-mode fiber endoscope

Abstract

Multi-mode fiber (MMF) endoscopes are extremely thin and have higher spatial resolution than conventional endoscopes; however, all current MMF endoscope designs require either that the MMF remain rigid during insertion and imaging or that the orientation of the MMF be known. This limits their possible medical applications. We describe an MMF endoscope design that allows the MMF to be arbitrarily bent as it is maneuvered to the target site prior to imaging. This is achieved by the addition of a partial reflector to the distal end of the MMF, which allows measurement of the mode coupling in the MMF using the reflected light arriving at the proximal end of the MMF. This measurement can be performed while the distal end of the endoscope is not directly accessible, as when the endoscope is being maneuvered. We simulate imaging through such a flexible MMF endoscope, where the MMF is step-index with 1588 spatial modes, and obtain an image even after the mode coupling matrix of the MMF is altered randomly, corresponding to an unknown bending of the MMF.

Characterizing Optical Fiber Transmission Matrices Using Metasurface Reflector Stacks for Lensless Imaging without Distal Access

“…In some cases this ambiguity reduces to an N -dimensional vector of sign errors (i.e. a vector ∈ {−1, 1} N ) for an N -pixel image, for example, when using MCF with low core-to-core coupling [17] or using an MMF that is near perfectly unitary (via the Autonne-Takagi matrix factorisation) [15]. In such cases, the ambiguity may be resolved by in situ optimisation assuming prior knowledge of the sample [17] or by treating the ambiguities as invariants of the fibre that can be measured in advance [15].…”

“…a vector ∈ {−1, 1} N ) for an N -pixel image, for example, when using MCF with low core-to-core coupling [17] or using an MMF that is near perfectly unitary (via the Autonne-Takagi matrix factorisation) [15]. In such cases, the ambiguity may be resolved by in situ optimisation assuming prior knowledge of the sample [17] or by treating the ambiguities as invariants of the fibre that can be measured in advance [15]. In practical cases, however, TMs are not unitary even in precision-manufactured MMF [18,19] because high-order modes, with more power concentrated in the fibre cladding, tend to exhibit power leakage at bends [20].…”

“…The complex-valued samples, representing amplitude and phase across two polarisations, can then be ordered to produce the vector x with 2M elements [3]. Similarly, if we consider a field, x ∈ C 2M , at the distal facet propagating in the reverse direction to become a field, y ∈ C 2M , at the proximal facet, these are related by the transpose of the fibre TM [15]:…”

“…Since we now only have access to the proximal facet of the fibre, we instead need to estimate the RM, C = A RA, based on the measured vectors (x, y). Given that we can characterise R in advance (Appendix A 3 a) and it will remain fixed throughout use, the goal is then to recover A based on measurements of C and prior characterisations of R, however, with only this information it is not in general possible to recover A unless the fibre TM is unitary [15,18]. We therefore propose a novel approach that uses several different reflectors, with PMs R n , n = 1..Q, and show that this enables unambiguous recovery of A in the more general non-unitary case.…”

Physical Layer Security in Multimode Fiber Optical Networks

“…The T M approach proved to be extremely helpful to overcome the challenge to use the MMF either as an imaging [22,23,24,25,26] or data transmitting tool [34,30] for compensating the fiber's light scrambling property. Besides experimental techniques to determine the T M , approaches to identify the T M via deep learning [35,36] or convex optimization algorithms [37] have already been shown.…”