Fig 1.
Schematic overview of the preparation protocol for human cochlear samples.
1) Extraction and fixation and decalcification, 2) μCT imaging, 3) dehydration and optical clearing 4) optional insertion of cochlear implant (CI), and 5) SLOT imaging.
Fig 2.
Illustration of optical clearing.
Left: Photograph of a decalcified cochlea sample in PBS. The cochlea is not transparent to visible light. Right: The same cochlea after optical clearing and immersion in MSBB. The optical clearing process has made the sample transparent to visible light, thereby making the text behind the cochlea legible. Two air bubbles can also be seen in the semicircular ducts which are outside of the area of interest. Air bubbles could otherwise lead to artifacts in reconstructed SLOT images.
Fig 3.
Optically cleared cochlea with cochlear implant Nucleus® CI 422 inserted.
The cochlea sample shown here is different from the one depicted in Fig 2.
Fig 4.
Schematic of the SLOT experimental setup.
A laser diode (LD) is coupled into the setup using a single mode optical fiber (SMF). At the end of the fiber, the light is collimated and its beam diameter is adjusted using a motorized zoom lens (ZL). The beam is focused on the rotational axis of the system and scanned across the sample in x-y-direction using a combination of 2-axis galvanometric scanning mirror (GM) and a telecentric Fθ-lens. The sample itself is inside a cuvette filled with MSBB and attached to the axis of rotation. Transmitted light is detected with a photodiode (PD). Fluorescence light is collected with a fiber bundle and filtered with a fluorescence emission filter (EF). Thus, scattered light is blocked before the signal is measured with a photomultiplier Tube (PMT). For each scan a digital projection image is formed in a computer and the sample is rotated for a full revolution in small angular steps.
Fig 5.
SLOT cross section images of the human cochlea.
Figures A to C show three orthogonal planes of the 3D dataset. The detection channels have been superposed: Autofluorescence (cyan) and extinction (red). The following anatomical structures can be clearly identified: scala tympani (ST), scala vestibuli and scala media (SV+SM), modiolus (M), basilar membrane (BM), the spiral ligament (SL), and osseous spiral lamina (OSL). Here, scala vestibuli (SV) and scala media (SM) are labeled as a unity, because Reissner’s membrane is not visible to distinguish between the two ducts and clearly localize them. The scale bar corresponds to 1 mm.
Fig 6.
μCT and SLOT images of the identical human cochlea sample.
The two data sets have been aligned after imaging: A μCT image in grey scale, B SLOT autofluorescence image in cyan, and C superposition of the two data sets for detailed comparison. It can be seen that one cochlear duct is not visible in the CT images which is most likely due to residual water inside the duct as a consequence of incorrect sample preparation. The scale bar corresponds to 1 mm.
Fig 7.
SLOT images of the human cochlea with cochlear implant electrode (Nucleus® CI 422) inserted.
Figures A to C show three orthogonal planes of the 3D dataset. The detection channels have been superposed: Autofluorescence (cyan) and extinction (red). The white arrows indicate radial artifacts due to the reflections at the metal surface. The scale bar corresponds to 1 mm.