Cochlear Duct Length Measurements in Computed Tomography and Magnetic Resonance Imaging Using Newly Developed Techniques

The Photon-Counting CT Enters the Field of Cochlear Implantation: Comparison to Angiography DynaCT and Conventional Multislice CT

INTRODUCTION

Cochlear duct length (CDL) measurement plays a role in the context of individualized cochlear implant (CI) surgery regarding an individualized selection and implantation of the CI electrode carrier and an efficient postoperative anatomy-based fitting process. The level of detail of the preoperative temporal bone CT scan depends on the imaging modality with major impact on CDL measurements and CI electrode contact position determination. The aim of this study was to evaluate the accuracy of perioperative CDL measurements and electrode contact determination in photon-counting CT (PCCT).

METHODS

Ten human fresh-frozen petrous bone specimens were examined with a first-generation PCCT. A clinically applicable radiation dose of 27.1 mGy was used. Scans were acquired before and after CI insertion. Postoperative measurement of the CDL was conducted using an otological planning software and 3D-curved multiplanar reconstruction. Investigation of electrode contact position was performed by two respective observers. Measurements were compared with a conventional multislice CT and to a high-resolution flat-panel volume CT with secondary reconstructions.

RESULTS

Pre- and postoperative CDL measurements in PCCT images showed no significant difference to high-resolution flat-panel volume CT. Postoperative CI electrode contact determination was also as precise as the flat-panel CT-based assessment. PCCT and flat-panel volume CT were equivalent concerning interobserver variability.

CONCLUSION

CDL measurement with PCCT was equivalent to flat-panel volume CT with secondary reconstructions. PCCT enabled highly precise postoperative CI electrode contact determination with substantial advantages over conventional multislice CT scanners.

Cochlear measurement in computed tomography and magnetic resonance imaging data sets by the Otoplan measurement tool: a retrospective comparative study

BACKGROUND

Using Otoplan software, it is possible to measure the cochlea before cochlear implant surgery. Until now, computed tomography (CT) of the cochlea has been necessary for this purpose. The aim of this study was to find out whether measuring the cochlea with magnetic resonance imaging (MRI) using Otoplan is possible with the same accuracy.

METHODS

The cochlea of 44 patients of the local cochlear implant centre was measured by Otoplan using high-resolution CT-bone and MRI images, and the determined lengths were compared.

RESULTS

No significant difference was found between the cochlear lengths measured, regardless of whether the length measurement was based on a CT or an MRI data set.

CONCLUSION

For the determination of cochlear length prior to cochlear implant surgery, MRI images are just as suitable as CT images, therefore CT is not mandatory for length measurement by Otoplan, which could reduce the patient's radiation exposure.

Enhancing cochlear duct length estimation by incorporating second-turn parameters

Estimating insertion depth, cochlear duct length (CDL), and other inner ear parameters is vital to optimizing cochlear implantation outcomes. Most current formulas use only the basal turn dimensions for CDL prediction. In this study, we investigated the importance of the second turn parameters in estimating CDL. Two experienced neuro-otologists blindly used segmentation software to measure (in mm) cochlear parameters, including basal turn diameter (A), basal turn width (B), second-turn diameter (A2), second-turn width (B2), CDL, first-turn length, and second-turn length (STL). These readings were taken from 33 computed tomography (CT) images of temporal bones from anatomically normal ears. We constructed regression models using A, B, A2, and B2 values fitted to CDL, two-turn length, and five-fold cross-validation to ensure model validity. CDL, A value, and STL were longer in males than in females. The mean B2/A2 ratio was 0.91 ± 0.06. Adding A2 and B2 values improved CDL prediction accuracy to 86.11%. Therefore, we propose a new formula for more accurate CDL estimation using A, B, A2, and B2 values. In conclusion, the findings of this study revealed a notable improvement in the prediction of two-turn length (2TL), and CDL by clinically appreciable margins upon adding A2 and B2 values to the prediction formulas.

Towards fully automated inner ear analysis with deep-learning-based joint segmentation and landmark detection framework

Automated analysis of the inner ear anatomy in radiological data instead of time-consuming manual assessment is a worthwhile goal that could facilitate preoperative planning and clinical research. We propose a framework encompassing joint semantic segmentation of the inner ear and anatomical landmark detection of helicotrema, oval and round window. A fully automated pipeline with a single, dual-headed volumetric 3D U-Net was implemented, trained and evaluated using manually labeled in-house datasets from cadaveric specimen ([Formula: see text]) and clinical practice ([Formula: see text]). The model robustness was further evaluated on three independent open-source datasets ([Formula: see text] scans) consisting of cadaveric specimen scans. For the in-house datasets, Dice scores of [Formula: see text], intersection-over-union scores of [Formula: see text] and average Hausdorff distances of [Formula: see text] and [Formula: see text] voxel units were achieved. The landmark localization task was performed automatically with an average localization error of [Formula: see text] voxel units. A robust, albeit reduced performance could be attained for the catalogue of three open-source datasets. Results of the ablation studies with 43 mono-parametric variations of the basal architecture and training protocol provided task-optimal parameters for both categories. Ablation studies against single-task variants of the basal architecture showed a clear performance benefit of coupling landmark localization with segmentation and a dataset-dependent performance impact on segmentation ability.

Comprehension of Cochlear Duct Length for Incomplete Partition Types

OBJECTIVE

Preoperative assessment of the cochlear duct length (CDL) and cochlear dimensions allows the selection of optimized implants. We aimed to evaluate the CDL measurements in incomplete partition (IP) defect patients and to create a reference to the literature.

METHODS

Forty-one patients with IP (13 IP I, 23 IP II, and 5 IP III) and 30 controls were included in the study. The standardized cochlear image showing the basal turn in the most expansive plane was reconstructed from temporal high-resolution computed tomography images. Cochlear duct length measured manually (CDL-M) was measured by points placed consecutively on the lateral wall of the cochlea. The defined equations for estimating CDL (CDL measured according to Schurzig et al formula [CDL-Ɵ], CDL measured according to Escudé et al formula [CDL-E], CDL measured according to Alexiades et al formula [CDL-A]) were calculated from the same images. Cochlear duct length mean values obtained by each method were compared for each IP type.

RESULTS

The longest CDL value was found in the control group, irrespective of the calculation method. Incomplete partition II cases had the most extended mean CDL among IP types. Incomplete partition III had the shortest CDL among all groups' CDL-M values. However, the mean CDL-M values of IP types I and III showed close results. There was no significant difference between the CDL-E and CDL-M values of the control group. Similarly, no significant difference was found between CDL-Ɵ and CDL-M values in IP type III cases. However, the results of other estimating formulations of all groups differed significantly from CDL-M values.

CONCLUSION

Cochlear duct length differences were detected between the control group and IP subtypes. These differences should be considered when choosing the appropriate electrode length. Because the results of formulas estimating CDL may differ from CDL-M in both control and IP cases, it would be more appropriate to use manual measurements in clinical practice.

Comparing accuracy of cochlear measurements on magnetic resonance imaging and computed tomography: A step towards radiation-free cochlear implantation

Objective

Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) are commonly employed in pre-operative evaluation for cochlear implant surgery. However, with a decrease in the age of implantation, even minor radiation exposure can cause detrimental effects on children over their lifetime. The current study compares different cochlear measurements from CT and MRI scans and evaluates the feasibility of using only an MRI scan for radiological evaluation before cochlear implantation.

Methods

A longitudinal observational study was conducted on 94 ears/47 children, employing CT and MRI scans. The CT and MRI scan measurements include, A value, B value, Cochlear duct length (CDL), two-turn cochlear length, alpha and beta angles to look for cochlear orientation. Cochlear nerve diameter was measured using MRI. The values were compared.

Results

The mean difference between measurements from CT and MRI scans for A value, B value, CDL, and two-turn cochlear length values was 0.567 ± 0.413 mm, 0.406 ± 0.368 mm, 2.365 ± 1.675 mm, and 2.063 ± 1.477 mm respectively without any significant difference. The alpha and beta angle measures were comparable, with no statistically significant difference.

Conclusion

The study suggests that MRI scans can be the only radiological investigation needed with no radiation risk and reduces the cost of cochlear implant program in the paediatric population. There is no significant difference between the measurements obtained from CT and MRI scans. However, observed discrepancies in cochlear measurements across different populations require regionally or race-specific standardized values to ensure accurate diagnosis and precision in cochlear implant surgery. This aspect must be addressed to ensure positive outcomes for patients.

Intra- and Interrater Reliability of CT- versus MRI-Based Cochlear Duct Length Measurement in Pediatric Cochlear Implant Candidates and Its Impact on Personalized Electrode Array Selection

Background: Radiological high-resolution computed tomography-based evaluation of cochlear implant candidates’ cochlear duct length (CDL) has become the method of choice for electrode array selection. The aim of the present study was to evaluate if MRI-based data match CT-based data and if this impacts on electrode array choice. Methods: Participants were 39 children. CDL, length at two turns, diameters, and height of the cochlea were determined via CT and MRI by three raters using tablet-based otosurgical planning software. Personalized electrode array length, angular insertion depth (AID), intra- and interrater differences, and reliability were calculated. Results: Mean intrarater difference of CT- versus MRI-based CDL was 0.528 ± 0.483 mm without significant differences. Individual length at two turns differed between 28.0 mm and 36.6 mm. Intrarater reliability between CT versus MRI measurements was high (intra-class correlation coefficient (ICC): 0.929–0.938). Selection of the optimal electrode array based on CT and MRI matched in 90.1% of cases. Mean AID was 629.5° based on the CT and 634.6° based on the MRI; this is not a significant difference. ICC of the mean interrater reliability was 0.887 for the CT-based evaluation and 0.82 for the MRI-based evaluation. Conclusion: MRI-based CDL measurement shows a low intrarater difference and a high interrater reliability and is therefore suitable for personalized electrode array selection.

[Measuring the cochlea using a tablet-based software package: influence of imaging modality and rater background]

BACKGROUND

Cochlear duct length (CDL) is subject to significant individual variation. In the context of cochlear implantation, adapting the electrode array length to the CDL is of potential interest, as it has been associated with improvements in both speech recognition and sound quality. Using a tablet-based software package, it is possible to measure CDL at the level of the organ of Corti (CDL_OC) to select appropriate electrode array lengths based on individual cochlear anatomy.

OBJECTIVE

To identify effects of imaging modality and rater background on CDL estimates.

METHODS

Magnetic resonance imaging (MRI) and flat-panel volume CT (fpVCT) scans of 10 patients (20 cochleae) were analyzed using the OTOPLAN software package (MED-EL, Innsbruck, Austria). Raters were an otorhinolaryngology (ORL) specialist, an ORL resident, and an audiologist. To analyze effects of rater background and imaging modality on CDL measurements, linear mixed models were constructed.

RESULTS

Measurements showed mean CDL_OC(fpVCT) = 36.69 ± 1.78 mm and CDL_OC(MRI) = 36.81 ± 1.87 mm. Analyses indicated no significant effect of rater background (F(2, 105) = 0.84; p = 0.437) on CDL estimates. Imaging modality, on the other hand, significantly affected CDL (F (1, 105) = 20.70; p < 0.001), whereby estimates obtained using MRI were 0.89 mm larger than those obtained using fpVCT.

CONCLUSION

No effect of rater background on CDL estimates could be identified, suggesting that comparable measurements could be obtained by personnel other than specially trained neurootologists. While imaging modality (fpVCT vs. MRI) did impact CDL results, the difference was small and of questionable clinical significance.

[Cochlear Implantation: Evaluation of Cochlear Duct Length (CDL)]

Personalized care in the context of cochlear implantation is becoming increasingly important. Choosing the right electrode could improve speech understanding. The measurement of the cochlear length plays an important role: preoperatively, in order to select a suitable electrode length; postoperatively, on the one hand to check the correct electrode position, on the other hand to enable anatomically based fitting of the electrode contacts. Of the various possible localizations of the CDL measurements within the cochlear turns, the one on the organ of Corti (CDL_OC) is the most frequently used and clinically most important. In the CDL measurement, a direct and indirect evaluation can be distinguished. There is also the possibility of reconstructing and measuring the CDL in 3D and calculating it mathematically, e.g. using spiral equations. In this context, measurements based on radiological imaging are gaining increasing importance. Therefore, if there is the possibility of performing higher-resolution imaging, this should be strived preoperatively in order to enable the most precise possible procedure and thus a good outcome. Otological planning software can help to create an interface between new findings regarding CDL measurement and higher-resolution imaging for an individualized cochlear implantation.