Effects of object-to-detector distance and beam energy on synchrotron radiation phase-contrast imaging of implanted cochleae.

OBJECTIVES: To demonstrate that synchrotron radiation phase-contrast imaging (SR-PCI) can be used to visualize the intrascalar structures in implanted human cochleae and to find the optimal combination of the parameters object-to-detector distance (ODD) and beam energy (E) for visualization. MATERIALS AND METHODS: Three cadaveric implanted human temporal bones underwent SR-PCI with varying combinations of parameters ODD (3, 2 and 1 m) and E (47, 60 and 72 keV). All images were then reconstructed to a three-dimensional (3D) stack of slices. The acquired 3D images were compared using contrast-to-noise ratios (CNRs) of the basilar membrane ( CNRBM ) and the electrode array (CNRE ) and the standard deviation of the beam streaks ( σS ). Postprocessing calculations were performed using Matlab (Version 2017b, MathWorks Inc., Natick, MA, U.S.A.) with a standard significance level p < 0.05 to determine the most optimal combination of parameters. RESULTS: SR-PCI with computed tomography reconstruction provided good visualization of the anatomical features of the implanted cochleae, specifically the exact location of the electrode with respect to the BM. A single-factor ANOVA revealed a significant difference of variance for both CNRE and CNRBM , but failed to show significance for σS . A two-sample t-test failed to show any significant difference between CNRE columns of (3 m, 72 keV) and (2 m, 60 keV). The CNRBM was significantly different only at two pairs of columns, when (1 m, 72 keV) was compared against (2 m, 72 keV) and (3 m, 72 keV). CONCLUSIONS: The results of this study show that SR-PCI is a viable method to visualize implanted human cochleae. SR-PCI is less invasive, less labour intensive and is associated with a much lower acquisition time compared to other methods for postimplantation imaging in humans, such as histological sectioning. We found that the optimal combination of E and ODD parameters was 72 keV and 2 m, respectively. These parameters resulted in high-contrast images of the electrode as well as all internal structures of the cochleae. LAY DESCRIPTION: Cochlear implants (CI) are currently the preferred method of treatment for hearing loss. Cochlear implantation surgery involves placement of a metallic, wire-shaped electrode inside the cochlea, the main organ of the human hearing system. Knowledge of the exact location of the electrode after implantation is beneficial in improving the extent of restored hearing. Common clinical imaging modalities such as computed-tomography (CT) are not ideal for providing such information, due to lack of resolution and streaking caused by the metallic electrode. Recent studies have developed algorithms to extract the electrode location from clinical computed-tomography images and have been validated using histology or micro computed-tomography (micro-CT). Synchrotron radiation phase contrast imaging (SR-PCI) is a high-resolution imaging technique used to visualize small structures in three dimensions. Recently, SR-PCI has been shown to be an alternative to histology or micro-CT for imaging the human cochlea. However, it has not been optimized for imaging implanted human cochleae. The main objective of the present work was to find the optimal organization of imaging parameters (i.e., object-to-detector distance and beam energy) for using SR-PCI to image implanted human cochleae. Three cadaveric human cochleae were imaged using five different combinations of imaging parameters at the Canadian Light Source Inc., Saskatoon, SK, Canada. The resulting images were compared both quantitatively and qualitatively. An optimal combination of parameters was found to produce high-contrast images of the both the CI electrode and all internal structures of the cochlea with minimal streaking. SR-PCI is therefore a viable alternative to histological or micro-CT studies for post-surgical imaging of implanted human cochleae.

An automated A-value measurement tool for accurate cochlear duct length estimation.

BACKGROUND: There has been renewed interest in the cochlear duct length (CDL) for preoperative cochlear implant electrode selection and postoperative generation of patient-specific frequency maps. The CDL can be estimated by measuring the A-value, which is defined as the length between the round window and the furthest point on the basal turn. Unfortunately, there is significant intra- and inter-observer variability when these measurements are made clinically. The objective of this study was to develop an automated A-value measurement algorithm to improve accuracy and eliminate observer variability. METHOD: Clinical and micro-CT images of 20 cadaveric cochleae specimens were acquired. The micro-CT of one sample was chosen as the atlas, and A-value fiducials were placed onto that image. Image registration (rigid affine and non-rigid B-spline) was applied between the atlas and the 19 remaining clinical CT images. The registration transform was applied to the A-value fiducials, and the A-value was then automatically calculated for each specimen. High resolution micro-CT images of the same 19 specimens were used to measure the gold standard A-values for comparison against the manual and automated methods. RESULTS: The registration algorithm had excellent qualitative overlap between the atlas and target images. The automated method eliminated the observer variability and the systematic underestimation by experts. Manual measurement of the A-value on clinical CT had a mean error of 9.5 ± 4.3% compared to micro-CT, and this improved to an error of 2.7 ± 2.1% using the automated algorithm. Both the automated and manual methods correlated significantly with the gold standard micro-CT A-values (r = 0.70, p < 0.01 and r = 0.69, p < 0.01, respectively). CONCLUSION: An automated A-value measurement tool using atlas-based registration methods was successfully developed and validated. The automated method eliminated the observer variability and improved accuracy as compared to manual measurements by experts. This open-source tool has the potential to benefit cochlear implant recipients in the future.

Intra- and Interobserver Variability of Cochlear Length Measurements in Clinical CT.

HYPOTHESIS: The cochlear A-value measurement exhibits significant inter- and intraobserver variability, and its accuracy is dependent on the visualization method in clinical computed tomography (CT) images of the cochlea. BACKGROUND: An accurate estimate of the cochlear duct length (CDL) can be used to determine electrode choice, and frequency map the cochlea based on the Greenwood equation. Studies have described estimating the CDL using a single A-value measurement, however the observer variability has not been assessed. METHODS: Clinical and micro-CT images of 20 cadaveric cochleae were acquired. Four specialists measured A-values on clinical CT images using both standard views and multiplanar reconstructed (MPR) views. Measurements were repeated to assess for intraobserver variability. Observer variabilities were evaluated using intra-class correlation and absolute differences. Accuracy was evaluated by comparison to the gold standard micro-CT images of the same specimens. RESULTS: Interobserver variability was good (average absolute difference: 0.77 ±â€Š0.42 mm) using standard views and fair (average absolute difference: 0.90 ±â€Š0.31 mm) using MPR views. Intraobserver variability had an average absolute difference of 0.31 ±â€Š0.09 mm for the standard views and 0.38 ±â€Š0.17 mm for the MPR views. MPR view measurements were more accurate than standard views, with average relative errors of 9.5 and 14.5%, respectively. CONCLUSION: There was significant observer variability in A-value measurements using both the standard and MPR views. Creating the MPR views increased variability between experts, however MPR views yielded more accurate results. Automated A-value measurement algorithms may help to reduce variability and increase accuracy in the future.