Resolution-enhanced multi-core fiber imaging learned on a digital twin for cancer diagnosis.

Abstract

Deep learning enables label-free all-optical biopsies and automated tissue classification. Endoscopic systems provide intraoperative diagnostics to deep tissue and speed up treatment without harmful tissue removal. However, conventional multi-core fiber (MCF) endoscopes suffer from low resolution and artifacts, which hinder tumor diagnostics. We introduce a method to enable unpixelated, high-resolution tumor imaging through a given MCF with a diameter of around 0.65mm and arbitrary core arrangement and inhomogeneous transmissivity. Image reconstruction is based on deep learning and the digital twin concept of the single-reference-based simulation with inhomogeneous optical properties of MCF and transfer learning on a small experimental dataset of biological tissue. The reference provided physical information about the MCF during the training processes. For the simulated data, hallucination caused by the MCF inhomogeneity was eliminated, and the averaged peak signal-to-noise ratio and structural similarity were increased from 11.2dB and 0.20 to 23.4dB and 0.74, respectively. By transfer learning, the metrics of independent test images experimentally acquired on glioblastoma tissue ex vivo can reach up to 31.6dB and 0.97 with 14fps computing speed. With the proposed approach, a single reference image was required in the pre-training stage and laborious acquisition of training data was bypassed. Validation on glioblastoma cryosections with transfer learning on only 50 image pairs showed the capability for high-resolution deep tissue retrieval and high clinical feasibility.