EXECUTIVE SUMMARY

Endoscopy has historically advanced by making imaging probes smaller, more flexible, and more clinically informative. Conventional fiber-based endoscopes rely heavily on coherent bundles of many individual glass fibers, often based on borosilicate or silica fiber architectures, where each fiber core functions as a discrete image element. This approach has enabled decades of clinical utility, but it also imposes a fundamental design tradeoff: as the bundle diameter decreases, the number of imaging cores decreases, pixelation becomes more pronounced, and spatial resolution becomes constrained by core packing density, inter-core spacing, and distal optics.

Multimode fiber offers a disruptive but still early-stage alternative. Instead of transmitting an image through thousands of physically separated cores, a single multimode fiber can transmit high spatial information density through many guided modes within one compact waveguide. With wavefront shaping and transmission-matrix calibration, multimode fiber can be transformed from an apparently scrambled light guide into a controllable imaging conduit. Early research has demonstrated that multimode-fiber-based endoscopy can support ultrathin, high-resolution imaging, including in vivo imaging with subcellular-scale potential, while reducing distal probe diameter and opening new procedural pathways. However, this should be viewed as an emerging platform that still requires further development in bend tolerance, clinical workflow integration, and regulatory validation before it can become a practical replacement for established fiber-bundle architectures.

The Limitations of Bundled Borosilicate Fiber Endoscopy

Coherent fiber bundles are conceptually simple: they preserve the relative position of thousands of individual fibers from the distal tip to the proximal image plane. Their strength is mechanical flexibility and direct image relay. Their weakness is that the imaging grid is physically fixed by the number, size, and spacing of individual fibers. The bundle therefore behaves like a hard-wired pixel array; any reduction in diameter tends to reduce the number of available cores, and the spaces between cores introduce dead zones that degrade resolution.

These limitations matter most in the applications where miniaturization is most valuable, such as pediatrics, brain imaging, and procedures within small blood vessels and the fine structures within the lungs. However, in traditional bundle designs, making the probe smaller can degrade the very image quality needed for confident diagnosis or therapy guidance. This creates a practical ceiling on how far bundled borosilicate architectures can scale into the next generation of endoscopy.