Imaging the ex-vivo human cochlea using 1.3-μm and 1.7-μm optical coherence tomography.

Significance: There is no clinical imaging method to visualize the soft tissues of the human cochlea, which are crucial for sound transduction and are damaged in sensorineural hearing loss. Although optical coherence tomography (OCT) has been effective in small animal models, we show for the first time that it can image through the full thickness of the ex-vivo human otic capsule and resolve cochlear microstructures despite increased scattering.

Aim: We aim to investigate whether OCT could image the cochlea through the otic capsule. We compared 1.7 and $1.3\mu m$ OCT to test if the reduced scattering at $1.7\mu m$ provided any appreciable advantage for imaging the cochleae.

Approach: OCT interferometers were built for both 1.3 and $1.7\mu m$ wavelengths, using identical sample and reference arm optics in both systems. Imaging was performed on two fixed human temporal bones with intact cochleae. The interferometers were designed to allow seamless switching between 1.3 and $1.7\mu m$ OCT without disrupting the temporal bone during imaging.

Results: We took volumetric OCT images at the base, apex, and hook regions of fixed ex-vivo human cochleae and compared the images taken at $1.3\mu m$ with those taken at $1.7\mu m$ . At both wavelengths, we could see through the otic capsule and identify cochlear structures. In some cases, $1.7\mu m$ OCT resulted in clearer images of the lateral wall, interior scala, and fine cochlear structures due to reduced multiple scattering at depth compared with $1.3\mu m$ .

Conclusions: We conclude that both $1.7\mu m$ and $1.3\mu m$ OCT can image through the human otic capsule, offering the potential for direct measurement of cochlear vibrometry or blood flow in living humans. Using $1.7\mu m$ light, we observed reduced multiple scattering in the otic capsule, leading to enhanced contrast of cochlear structures compared with $1.3\mu m$ . However, these improvements were marginal and came with trade-offs.