Automatic segmentation of intracochlear anatomy in conventional CT

Morphologic Analysis of the Scala Tympani Using Synchrotron: Implications for Cochlear Implantation

OBJECTIVES

To use synchrotron radiation phase-contrast imaging (SR-PCI) to visualize and measure the morphology of the entire cochlear scala tympani (ST) and assess cochlear implant (CI) electrode trajectories.

METHODS

SR-PCI images were used to obtain geometric measurements of the cochlear scalar diameter and area at 5-degree increments in 35 unimplanted and three implanted fixed human cadaveric cochleae.

RESULTS

The cross-sectional diameter and area of the cochlea were found to decrease from the base to the apex. This study represents a wide variability in cochlear morphology and suggests that even in the smallest cochlea, the ST can accommodate a 0.4 mm diameter electrode up to 720°. Additionally, all lateral wall array trajectories were within the anatomically accommodating insertion zone.

CONCLUSION

This is the first study to use SR-PCI to visualize and quantify the entire ST morphology, from the round window to the apical tip, and assess the post-operative trajectory of electrodes. These high-resolution anatomical measurements can be used to inform the angular insertion depth that can be accommodated in CI patients, accounting for anatomical variability.

LEVEL OF EVIDENCE

N/A. Laryngoscope, 134:2889-2897, 2024.

Improving rehabilitation of deaf patients by advanced imaging before cochlear implantation

INTRODUCTION

Cochlear implants have advanced the management of severe to profound deafness. However, there is a strong disparity in hearing performance after implantation from one patient to another. Moreover, there are several advanced kinds of imaging assessment before cochlear implantation. Microstructural white fiber degeneration can be studied with Diffusion weighted MRI (DWI) or tractography of the central auditory pathways. Functional MRI (fMRI) allows us to evaluate brain function, and CT or MRI segmentation to better detect inner ear anomalies.

OBJECTIVE

This literature review aims to evaluate how helpful pre-implantation anatomic imaging can be to predict hearing rehabilitation outcomes in deaf patients. These techniques include DWI and fMRI of the central auditory pathways, and automated labyrinth segmentation by CT scan, cone beam CT and MRI.

DESIGN

This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies were selected by searching in PubMed and by checking the reference lists of relevant articles. Inclusion criteria were adults over 18, with unilateral or bilateral hearing loss, who had DWI acquisition or fMRI or CT/ Cone Beam CT/ MRI image segmentation.

RESULTS

After reviewing 172 articles, we finally included 51. Studies on DWI showed changes in the central auditory pathways affecting the white matter, extending to the primary and non-primary auditory cortices, even in sudden and mild hearing impairment. Hearing loss patients show a reorganization of brain activity in various areas, such as the auditory and visual cortices, as well as regions involved in language and emotions, according to fMRI studies. Deep Learning's automatic segmentation produces the best CT segmentation in just a few seconds. MRI segmentation is mainly used to evaluate fluid space of the inner ear and determine the presence of an endolymphatic hydrops.

CONCLUSION

Before cochlear implantation, a DWI with tractography can evaluate the central auditory pathways up to the primary and non-primary auditory cortices. This data is then used to generate predictions on the auditory rehabilitation of patients. A CT segmentation with systematic 3D reconstruction allow a better evaluation of cochlear malformations and predictable difficulties during surgery.

Parameterisation and Prediction of Intra-canal Cochlear Structures

Accurate 3D models of the cochlea are useful tools for research in the relationship between the electrode array and nerve fibres. The internal geometry of the cochlear canal plays an important role in understanding and quantifying that relationship. Predicting the location and shapes of the geometry is done by measuring histologic sections and fitting equations that can be used to predict parameters that fully define the geometry. A parameter sensitivity analysis is employed to prove that the size and location of the spiral lamina are the characteristics that most influence current distribution along target nerve fibres. The proposed landmark prediction method more accurately predicts the location of the points defining the spiral lamina in the apical region of the cochlea than methods used in previous modelling attempts. Thus, this technique can be used to generate 2D geometries that can be expanded to 3D models when high-resolution imaging is not available.

Landmark-based registration of a cochlear model to a human cochlea using conventional CT scans

Cochlear implants can provide an advanced treatment option to restore hearing. In standard pre-implant procedures, many factors are already considered, but it seems that not all underlying factors have been identified yet. One reason is the low quality of the conventional computed tomography images taken before implantation, making it difficult to assess these parameters. A novel method is presented that uses the Pietsch Model, a well-established model of the human cochlea, as well as landmark-based registration to address these challenges. Different landmark numbers and placements are investigated by visually comparing the mean error per landmark and the registrations' results. The landmarks on the first cochlear turn and the apex are difficult to discern on a low-resolution CT scan. It was possible to achieve a mean error markedly smaller than the image resolution while achieving a good visual fit on a cochlear segment and directly in the conventional computed tomography image. The employed cochlear model adjusts image resolution problems, while the effort of setting landmarks is markedly less than the segmentation of the whole cochlea. As a next step, the specific parameters of the patient could be extracted from the adapted model, which enables a more personalized implantation with a presumably better outcome.

Cochlear Implant Fold Detection in Intra-operative CT Using Weakly Supervised Multi-task Deep Learning

In cochlear implant (CI) procedures, an electrode array is surgically inserted into the cochlea. The electrodes are used to stimulate the auditory nerve and restore hearing sensation for the recipient. If the array folds inside the cochlea during the insertion procedure, it can lead to trauma, damage to the residual hearing, and poor hearing restoration. Intraoperative detection of such a case can allow a surgeon to perform reimplantation. However, this intraoperative detection requires experience and electrophysiological tests sometimes fail to detect an array folding. Due to the low incidence of array folding, we generated a dataset of CT images with folded synthetic electrode arrays with realistic metal artifact. The dataset was used to train a multitask custom 3D-UNet model for array fold detection. We tested the trained model on real post-operative CTs (7 with folded arrays and 200 without). Our model could correctly classify all the fold-over cases while misclassifying only 3 non fold-over cases. Therefore, the model is a promising option for array fold detection.

Evaluation of CI electrode position from imaging: comparison of an automated technique with the established manual method

BACKGROUND

A manual evaluation of the CI electrode position from CT and DVT scans may be affected by diagnostic errors due to cognitive biases. The aim of this study was to compare the CI electrode localization using an automated method (image-guided cochlear implant programming, IGCIP) with the clinically established manual method.

METHODS

This prospective experimental study was conducted on a dataset comprising N=50 subjects undergoing cochlear implantation with a Nucleus® CI532 or CI632 Slim Modiolar electrode. Scalar localization, electrode-to-modiolar axis distances (EMD) and angular insertion depth (aDOI) were compared between the automated IGCIP tool and the manual method. Two raters made the manual measurements, and the interrater reliability (±1.96·SD) was determined as the reference for the method comparison. The method comparison was performed using a correlation analysis and a Bland-Altman analysis.

RESULTS

Concerning the scalar localization, all electrodes were localized both manually and automatically in the scala tympani. The interrater differences ranged between ±0.2 mm (EMD) and ±10° (aDOI). There was a bias between the automatic and manual method in measuring both localization parameters, which on the one hand was smaller than the interrater variations. On the other hand, this bias depended on the magnitude of the EMD respectively aDOI. A post-hoc analysis revealed that the deviations between the methods were likely due to a different selection of mid-modiolar axis.

CONCLUSIONS

The IGCIP is a promising tool for automated processing of CT and DVT scans and has useful functionality such as being able to segment the cochlear using post-operative scans. When measuring EMD, the IGCIP tool is superior to the manual method because the smallest possible distance to the axis is determined depending on the cochlear turn, whereas the manual method selects the helicotrema as the reference point rigidly. Functionality to deal with motion artifacts and measurements of aDOI according to the consensus approach are necessary, otherwise the IGCIP is not unrestrictedly ready for clinical use.

Automatic electrode scalar location assessment after cochlear implantation using a novel imaging software

As of today, image-based assessment of cochlear implant electrode array location is not part of the clinical routine. Low resolution and contrast of computer tomography (CT) imaging, as well as electrode array artefacts, prevent visibility of intracochlear structures and result in low accuracy in determining location of the electrode array. Further, trauma assessment based on clinical-CT images requires a uniform image-based trauma scaling. Goal of this study was to evaluate the accuracy of a novel imaging software to detect electrode scalar location. Six cadaveric temporal bones were implanted with Advanced Bionics SlimJ and Mid-Scala electrode arrays. Clinical-CT scans were taken pre- and postoperatively. In addition, micro-CTs were taken post-operatively for validation. The electrode scalar location rating done by the software was compared to the rating of two experienced otosurgeons and the micro-CT images. A 3-step electrode scalar location grading scale (0 = electrode in scala tympani, 1 = interaction of electrode with basilar membrane/osseous spiral lamina, 2 = translocation of electrode into scala vestibuli) was introduced for the assessment. The software showed a high sensitivity of 100% and a specificity of 98.7% for rating the electrode location. The correlation between rating methods was strong (kappa > 0.890). The software gives a fast and reliable method of evaluating electrode scalar location for cone beam CT scans. The introduced electrode location grading scale was adapted for assessing clinical CT images.

Diagnosis, Treatment, and Management of Otitis Media with Artificial Intelligence

A common infectious disease, otitis media (OM) has a low rate of early diagnosis, which significantly increases the difficulty of treating the disease and the likelihood of serious complications developing including hearing loss, speech impairment, and even intracranial infection. Several areas of healthcare have shown great promise in the application of artificial intelligence (AI) systems, such as the accurate detection of diseases, the automated interpretation of images, and the prediction of patient outcomes. Several articles have reported some machine learning (ML) algorithms such as ResNet, InceptionV3 and Unet, were applied to the diagnosis of OM successfully. The use of these techniques in the OM is still in its infancy, but their potential is enormous. We present in this review important concepts related to ML and AI, describe how these technologies are currently being applied to diagnosing, treating, and managing OM, and discuss the challenges associated with developing AI-assisted OM technologies in the future.

Validation of active shape model techniques for intracochlear anatomy segmentation in computed tomography images

Purpose

Cochlear implants (CIs) have been shown to be highly effective restorative devices for patients suffering from severe-to-profound hearing loss. Hearing outcomes with CIs depend on electrode positions with respect to intracochlear anatomy. Intracochlear anatomy can only be directly visualized using high-resolution modalities, such as micro-computed tomography (μ CT ), which cannot be used in vivo. However, active shape models (ASM) have been shown to be robust and effective for segmenting intracochlear anatomy in large scale datasets of patient computed tomographies (CTs). We present an extended dataset of μ CT specimens and aim to evaluate the ASM's performance more comprehensively than has been previously possible.

Approach

Using a dataset of 16 manually segmented cochlea specimens on μ CTs , we found parameters that optimize mean CT segmentation performance and then evaluate the effect of library size on the ASM. The optimized ASM was further evaluated on a clinical dataset of 134 CT images to assess method reliability.

Results

Optimized parameters lead to mean CT segmentation performance to 0.36 mm point-to-point error, 0.10 mm surface error, and 0.83 Dice score. Larger library sizes provide diminishing returns on segmentation performance and total variance captured by the ASM. We found our method to be clinically reliable with the main performance limitation that was found to be the candidate search process rather than model representation.

Conclusions

We have presented a comprehensive validation of the ASM for use in intracochlear anatomy segmentation. These results are critical to understand the limitations of the method for clinical use and for future development.

Dynamic Behavior and Insertional Forces of a Precurved Electrode Using the Pull-Back Technique in a Fresh Microdissected Cochlea

HYPOTHESIS

This study evaluated the utility of the pull-back technique in improving perimodiolar positioning of a precurved cochlear implant (CI) electrode array (EA) with simultaneous insertion force profile measurement and direct observation of dynamic EA behavior.

BACKGROUND

Precurved EAs with perimodiolar positioning have improved outcomes compared with straight EAs because of lowered charge requirements for stimulation and decreased spread of excitation. The safety and efficacy of the pull-back technique in further improving perimodiolar positioning and its associated force profile have not been adequately demonstrated.

METHODS

The bone overlying the scala vestibuli was removed in 15 fresh cadaveric temporal bones, leaving the scala tympani unviolated. Robotic insertions of EAs were performed with simultaneous force measurement and video recording. Force profiles were obtained during standard insertion, overinsertion, and pull-back. Postinsertion CT scans were obtained during each of the three conditions, enabling automatic segmentation and calculation of angular insertion depth, mean perimodiolar distance ( Mavg ), and cochlear duct length.

RESULTS

Overinsertion did not result in significantly higher peak forces than standard insertion (mean [SD], 0.18 [0.06] and 0.14 [0.08] N; p = 0.18). Six temporal bones (40%) demonstrated visibly improved perimodiolar positioning after the protocol, whereas none worsened. Mavg significantly improved after the pull-back technique compared with standard insertion (mean [SD], 0.34 [0.07] and 0.41 [0.10] mm; p < 0.01).

CONCLUSIONS

The pull-back technique was not associated with significantly higher insertional forces compared with standard insertion. This technique was associated with significant improvement in perimodiolar positioning, both visually and quantitatively, independent of cochlear size.

Deep learning-based approach for the automatic segmentation of adult and pediatric temporal bone computed tomography images

Background

Automatic segmentation of temporal bone computed tomography (CT) images is fundamental to image-guided otologic surgery and the intelligent analysis of CT images in the field of otology. This study was conducted to test a convolutional neural network (CNN) model that can automatically segment almost all temporal bone anatomy structures in adult and pediatric CT images.

Methods

A dataset comprising 80 annotated CT volumes was collected, of which 40 samples were obtained from adults and 40 from children. A further 60 annotated CT volumes (30 from adults and 30 from children) were used to train the model. The remaining 20 annotated CT volumes were employed to determine the model's generalizability for automatic segmentation. Finally, the Dice coefficient (DC) and average symmetric surface distance (ASSD) were utilized as metrics to evaluate the performance of the CNN model. Two independent-sample t-tests were used to compare the test set results of adults and children.

Results

In the adult test set, the mean DC values of all the structures ranged from 0.714 to 0.912, and the ASSD values were less than 0.24 mm for 11 structures. In the pediatric test set, the mean DC values of all the structures ranged from 0.658 to 0.915, and the ASSD values were less than 0.18 mm for 11 structures. There was no statistically significant difference between the adult and child test sets in most temporal bone structures.

Conclusions

Our CNN model shows excellent automatic segmentation performance and good generalizability for both adult and pediatric temporal bone CT images, which can help to advance otologist education, intelligent imaging diagnosis, surgery simulation, application of augmented reality, and preoperative planning for image-guided otology surgery.

Electrode array positioning after cochlear reimplantation from single manufacturer

OBJECTIVE

To investigate whether revision surgery with the same device results in a change in three key indicators of electrode positioning: scalar location, mean modiolar distance (M¯), and angular insertion depth (AID).

METHODS

Retrospective analysis of a cochlear implant database at a university-based tertiary medical center. Intra-operative CT scans were obtained after initial and revision implantation. Electrode array (EA) position was calculated using auto-segmentation techniques. Initial and revision scalar location, M¯, and AID were compared.

RESULTS

Mean change in M¯ for all ears was -0.07 mm (SD 0.24 mm; P = 0.16). The mean change in AID for all ears was -5° (SD 67°; P = 0.72). Three initial implantations with pre-curved EAs resulted in a translocation from Scala Tympani (ST) to Scala Vestibuli (SV). Two remained translocated after revision, while one was corrected when revised with a straight EA. An additional five translocations occurred after revision.

CONCLUSIONS

In this study examining revision cochlear implantation from a single manufacturer, we demonstrated no significant change in key indicators of EA positioning, even when revising with a different style of electrode. However, the revision EA is not necessarily confined by the initial trajectory and there may be an increased risk of translocation.

A Web-Based Automated Image Processing Research Platform for Cochlear Implantation-Related Studies

The robust delineation of the cochlea and its inner structures combined with the detection of the electrode of a cochlear implant within these structures is essential for envisaging a safer, more individualized, routine image-guided cochlear implant therapy. We present Nautilus-a web-based research platform for automated pre- and post-implantation cochlear analysis. Nautilus delineates cochlear structures from pre-operative clinical CT images by combining deep learning and Bayesian inference approaches. It enables the extraction of electrode locations from a post-operative CT image using convolutional neural networks and geometrical inference. By fusing pre- and post-operative images, Nautilus is able to provide a set of personalized pre- and post-operative metrics that can serve the exploration of clinically relevant questions in cochlear implantation therapy. In addition, Nautilus embeds a self-assessment module providing a confidence rating on the outputs of its pipeline. We present a detailed accuracy and robustness analyses of the tool on a carefully designed dataset. The results of these analyses provide legitimate grounds for envisaging the implementation of image-guided cochlear implant practices into routine clinical workflows.

Computed-Tomography Estimates of Interaural Mismatch in Insertion Depth and Scalar Location in Bilateral Cochlear-Implant Users

HYPOTHESIS

Bilateral cochlear-implant (BI-CI) users will have a range of interaural insertion-depth mismatch because of different array placement or characteristics. Mismatch will be larger for electrodes located near the apex or outside scala tympani, or for arrays that are a mix of precurved and straight types.

BACKGROUND

Brainstem superior olivary-complex neurons are exquisitely sensitive to interaural-difference cues for sound localization. Because these neurons rely on interaurally place-of-stimulation-matched inputs, interaural insertion-depth or scalar-location differences for BI-CI users could cause interaural place-of-stimulation mismatch that impairs binaural abilities.

METHODS

Insertion depths and scalar locations were calculated from temporal-bone computed-tomography scans for 107 BI-CI users (27 Advanced Bionics, 62 Cochlear, 18 MED-EL).

RESULTS

Median interaural insertion-depth mismatch was 23.4 degrees or 1.3 mm. Mismatch in the estimated clinically relevant range expected to impair binaural processing (>75 degrees or 3 mm) occurred for 13 to 19% of electrode pairs overall, and for at least three electrode pairs for 23 to 37% of subjects. There was a significant three-way interaction between insertion depth, scalar location, and array type. Interaural insertion-depth mismatch was largest for apical electrodes, for electrode pairs in two different scala, and for arrays that were both-precurved.

CONCLUSION

Average BI-CI interaural insertion-depth mismatch was small; however, large interaural insertion-depth mismatch-with the potential to degrade spatial hearing-occurred frequently enough to warrant attention. For new BICI users, improved surgical techniques to avoid interaural insertion-depth and scalar mismatch are recommended. For existing BI-CI users with interaural insertion-depth mismatch, interaural alignment of clinical frequency tables might reduce negative spatial-hearing consequences.

Best Fit 3D Basilar Membrane Reconstruction to Routinely Assess the Scalar Position of the Electrode Array after Cochlear Implantation

The scalar position of the electrode array is assumed to be associated with auditory performance after cochlear implantation. We propose a new method that can be routinely applied in clinical practice to assess the position of an electrode array. Ten basilar membrane templates were generated using micro-computed tomography (micro-CT), based on the dimensions of 100 cochleae. Five surgeons were blinded to determine the position of the electrode array in 30 cadaveric cochleae. The procedure consisted of selecting the appropriate template based on cochlear dimensions, merging the electrode array reconstruction with the template using four landmarks, determining the position of the array according to the template position, and comparing the results obtained to histology data. The time taken to analyze each implanted cochlea was approximately 12 min. We found that, according to histology, surgeons were in almost perfect agreement when determining an electrode translocated to the scala vestibuli with the perimodiolar MidScala array (Fleiss’ kappa (κ) = 0.82), and in moderate agreement when using the lateral wall EVO array (κ = 0.42). Our data indicate that an adapted basilar membrane template can be used as a rapid and reproducible method to assess the position of the electrode array after cochlear implantation.

Interaural Place-of-Stimulation Mismatch Estimates Using CT Scans and Binaural Perception, But Not Pitch, Are Consistent in Cochlear-Implant Users

Bilateral cochlear implants (BI-CIs) or a CI for single-sided deafness (SSD-CI; one normally functioning acoustic ear) can partially restore spatial-hearing abilities, including sound localization and speech understanding in noise. For these populations, however, interaural place-of-stimulation mismatch can occur and thus diminish binaural sensitivity that relies on interaurally frequency-matched neurons. This study examined whether plasticity-reorganization of central neural pathways over time-can compensate for peripheral interaural place mismatch. We hypothesized differential plasticity across two systems: none for binaural processing but adaptation for pitch perception toward frequencies delivered by the specific electrodes. Interaural place mismatch was evaluated in 19 BI-CI and 23 SSD-CI human subjects (both sexes) using binaural processing (interaural-time-difference discrimination with simultaneous bilateral stimulation), pitch perception (pitch ranking for single electrodes or acoustic tones with sequential bilateral stimulation), and physical electrode-location estimates from computed-tomography (CT) scans. On average, CT scans revealed relatively little BI-CI interaural place mismatch (26° insertion-angle mismatch) but a relatively large SSD-CI mismatch, particularly at low frequencies (166° for an electrode tuned to 300 Hz, decreasing to 14° at 7000 Hz). For BI-CI subjects, the three metrics were in agreement because there was little mismatch. For SSD-CI subjects, binaural and CT measurements were in agreement, suggesting little binaural-system plasticity induced by mismatch. The pitch measurements disagreed with binaural and CT measurements, suggesting place-pitch plasticity or a procedural bias. These results suggest that reducing interaural place mismatch and potentially improving binaural processing by reprogramming the CI frequency allocation would be better done using CT-scan than pitch information.SIGNIFICANCE STATEMENT Electrode-array placement for cochlear implants (bionic prostheses that partially restore hearing) does not explicitly align neural representations of frequency information. The resulting interaural place-of-stimulation mismatch can diminish spatial-hearing abilities. In this study, adults with two cochlear implants showed reasonable interaural alignment, whereas those with one cochlear implant but normal hearing in the other ear often showed mismatch. In cases of mismatch, binaural sensitivity was best when the same cochlear locations were stimulated in both ears, suggesting that binaural brainstem pathways do not experience plasticity to compensate for mismatch. In contrast, interaurally pitch-matched electrodes deviated from cochlear-location estimates and did not optimize binaural sensitivity. Clinical correction of interaural place mismatch using binaural or computed-tomography (but not pitch) information may improve spatial-hearing benefits.

Hybrid active shape and deep learning method for the accurate and robust segmentation of the intracochlear anatomy in clinical head CT and CBCT images

Purpose: Robust and accurate segmentation methods for the intracochlear anatomy (ICA) are a critical step in the image-guided cochlear implant programming process. We have proposed an active shape model (ASM)-based method and a deep learning (DL)-based method for this task, and we have observed that the DL method tends to be more accurate than the ASM method while the ASM method tends to be more robust. Approach: We propose a DL-based U-Net-like architecture that incorporates ASM segmentation into the network. A quantitative analysis is performed on a dataset that consists of 11 cochlea specimens for which a segmentation ground truth is available. To qualitatively evaluate the robustness of the method, an experienced expert is asked to visually inspect and grade the segmentation results on a clinical dataset made of 138 image volumes acquired with conventional CT scanners and of 39 image volumes acquired with cone beam CT (CBCT) scanners. Finally, we compare training the network (1) first with the ASM results, and then fine-tuning it with the ground truth segmentation and (2) directly with the specimens with ground truth segmentation. Results: Quantitative and qualitative results show that the proposed method increases substantially the robustness of the DL method while having only a minor detrimental effect (though not significant) on its accuracy. Expert evaluation of the clinical dataset shows that by incorporating the ASM segmentation into the DL network, the proportion of good segmentation cases increases from 60/177 to 119/177 when training only with the specimens and increases from 129/177 to 151/177 when pretraining with the ASM results. Conclusions: A hybrid ASM and DL-based segmentation method is proposed to segment the ICA in CT and CBCT images. Our results show that combining DL and ASM methods leads to a solution that is both robust and accurate.

Fully automated segmentation in temporal bone CT with neural network: a preliminary assessment study

BACKGROUND

Segmentation of important structures in temporal bone CT is the basis of image-guided otologic surgery. Manual segmentation of temporal bone CT is time- consuming and laborious. We assessed the feasibility and generalization ability of a proposed deep learning model for automated segmentation of critical structures in temporal bone CT scans.

METHODS

Thirty-nine temporal bone CT volumes including 58 ears were divided into normal (n = 20) and abnormal groups (n = 38). Ossicular chain disruption (n = 10), facial nerve covering vestibular window (n = 10), and Mondini dysplasia (n = 18) were included in abnormal group. All facial nerves, auditory ossicles, and labyrinths of the normal group were manually segmented. For the abnormal group, aberrant structures were manually segmented. Temporal bone CT data were imported into the network in unmarked form. The Dice coefficient (DC) and average symmetric surface distance (ASSD) were used to evaluate the accuracy of automatic segmentation.

RESULTS

In the normal group, the mean values of DC and ASSD were respectively 0.703, and 0.250 mm for the facial nerve; 0.910, and 0.081 mm for the labyrinth; and 0.855, and 0.107 mm for the ossicles. In the abnormal group, the mean values of DC and ASSD were respectively 0.506, and 1.049 mm for the malformed facial nerve; 0.775, and 0.298 mm for the deformed labyrinth; and 0.698, and 1.385 mm for the aberrant ossicles.

CONCLUSIONS

The proposed model has good generalization ability, which highlights the promise of this approach for otologist education, disease diagnosis, and preoperative planning for image-guided otology surgery.

Automatic Segmentation of Intracochlear Anatomy in MR Images Using a Weighted Active Shape Model

There is evidence that cochlear MR signal intensity may be useful in prognosticating the risk of hearing loss after middle cranial fossa (MCF) resection of acoustic neuroma (AN), but the manual segmentation of this structure is difficult and prone to error. This hampers both large-scale retrospective studies and routine clinical use of this information. To address this issue, we present a fully automatic method that permits the segmentation of the intra-cochlear anatomy in MR images, which uses a weighted active shape model we have developed and validated to segment the intra-cochlear anatomy in CT images. We take advantage of a dataset for which both CT and MR images are available to validate our method on 132 ears in 66 high-resolution T2-weighted MR images. Using the CT segmentation as ground truth, we achieve a mean Dice (DSC) value of 0.81 and 0.79 for the scala tympani (ST) and the scala vestibuli (SV), which are the two main intracochlear structures.Clinical Relevance- The proposed method is accurate and fully automated for MR image segmentation. It can be used to support large retrospective studies that explore relations between MR signal in preoperative images and outcomes. It can also facilitate the routine and clinical use of this information.

Estimating the Pose of a Guinea-pig Cochlea Without Medical Imaging

HYPOTHESIS

The pose (i.e., position and orientation) of a guinea-pig cochlea can be accurately estimated using externally observable features, without requiring computed-tomography (CT) scans.

BACKGROUND

Guinea pigs are frequently used in otologic research as animal models of cochlear-implant surgery. In robot-assisted surgical insertion of cochlear-implant electrode arrays, knowing the cochlea pose is required. A preoperative CT scan of the guinea-pig anatomy can be labeled and registered to the surgical system, however, this process can be expensive and time consuming.

METHODS

Anatomical features from both sides of 11 guinea-pig CT scans were labeled and registered, forming sets. Using a groupwise point-set registration algorithm, errors in cochlea position and modiolar-axis orientation were estimated for 11 iterations of registration where each feature set was used as a hold-out set containing a reduced number of features that could all be touched by a motion-tracking probe intraoperatively. The method was validated on 2000 simulated guinea-pig cochleae and six physical guinea-pig-skull cochleae.

RESULTS

Validation on simulated cochleae resulted in cochlea-position estimates with a maximum error of 0.43 mm and modiolar-axis orientation estimates with a maximum error of 8.1 degrees for 96.7% of cochleae. Physical validation resulted in cochlea-position estimates with a maximum error of 0.80 mm and modiolar-axis orientation estimates with a maximum error of 12.4 degrees.

CONCLUSIONS

This work enables researchers conducting robot-assisted surgical insertions of cochlear-implant electrode arrays using a guinea-pig animal model to estimate the pose of a guinea-pig cochlea by locating six externally observable features on the guinea pig, without the need for CT scans.

Influence of electrode to cochlear duct length ratio on post-operative speech understanding outcomes

OBJECTIVE

To assess whether the pre-operative electrode to cochlear duct length ratio (ECDLR), is associated with post-operative speech recognition outcomes.

STUDY DESIGN

A retrospective chart review study.

SETTING

Tertiary referral center.

PATIENTS

The study included sixty-one adult CI recipients with a pre-operative computed tomography scan and a speech recognition test 12 months after implantation.

INTERVENTIONS

The average of two raters' cochlear duct length (CDL) measurements and the length of the recipient's cochlear implant electrode array formed the basis for the electrode-to-cochlear duct length ratio (ECLDR). Speech recognition tests were compared as a function of ECDLR and electrode array length itself.

MAIN OUTCOME MEASURES

The relationship between ECDLR and percent correct on speech recognition tests.

RESULTS

A second order polynomial regression relating ECDLR to percent correct on the CNC words speech recognition test was statistically significant, as was a fourth order polynomial regression for the AzBio Quiet test. In contrast, there was no statistically significant relationship between speech recognition scores and electrode array length.

CONCLUSIONS

ECDLR values can be statistically associated to speech-recognition outcomes. However, these ECDLR values cannot be predicted by the electrode length alone, and must include a measure of CDL.

Atlas-based Segmentation of Intracochlear Anatomy in Metal Artifact Affected CT Images of the Ear with Co-trained Deep Neural Networks

We propose an atlas-based method to segment the intracochlear anatomy (ICA) in the post-implantation CT (Post-CT) images of cochlear implant (CI) recipients that preserves the point-to-point correspondence between the meshes in the atlas and the segmented volumes. To solve this problem, which is challenging because of the strong artifacts produced by the implant, we use a pair of co-trained deep networks that generate dense deformation fields (DDFs) in opposite directions. One network is tasked with registering an atlas image to the Post-CT images and the other network is tasked with registering the Post-CT images to the atlas image. The networks are trained using loss functions based on voxel-wise labels, image content, fiducial registration error, and cycle-consistency constraint. The segmentation of the ICA in the Post-CT images is subsequently obtained by transferring the predefined segmentation meshes of the ICA in the atlas image to the Post-CT images using the corresponding DDFs generated by the trained registration networks. Our model can learn the underlying geometric features of the ICA even though they are obscured by the metal artifacts. We show that our end-to-end network produces results that are comparable to the current state of the art (SOTA) that relies on a two-steps approach that first uses conditional generative adversarial networks to synthesize artifact-free images from the Post-CT images and then uses an active shape model-based method to segment the ICA in the synthetic images. Our method requires a fraction of the time needed by the SOTA, which is important for end-user acceptance.

Clinical Translation of an Insertion Tool for Minimally Invasive Cochlear Implant Surgery

The objective of this paper is to describe the development of a minimally invasive cochlear implant surgery (MICIS) electrode array insertion tool concept to enable clinical translation. First, analysis of the geometric parameters of potential MICIS patients (N = 97) was performed to inform tool design, inform MICIS phantom model design, and provide further insight into MICIS candidacy. Design changes were made to the insertion tool based on clinical requirements and parameter analysis results. A MICIS phantom testing model was built to evaluate insertion force profiles in a clinically realistic manner, and the new tool design was evaluated in the model and in cadavers to test clinical viability. Finally, after regulatory approval, the tool was used for the first time in a clinical case. Results of this work included first, in the parameter analysis, approximately 20% of the population was not considered viable MICIS candidates. Additionally, one 3D printed tool could accommodate all viable candidates with polyimide sheath length adjustments accounting for interpatient variation. The insertion tool design was miniaturized out of clinical necessity and a disassembly method, necessary for removal around the cochlear implant, was developed and tested. Phantom model testing revealed that the force profile of the insertion tool was similar to that of traditional forceps insertion. Cadaver testing demonstrated that all clinical requirements (including complete disassembly) were achieved with the tool, and the new tool enabled 15% deeper insertions compared to the forceps approach. Finally, and most importantly, the tool helped achieve a full insertion in its first MICIS clinical case. In conclusion, the new insertion tool provides a clinically viable solution to one of the most difficult aspects of MICIS.

An Approach for Individualized Cochlear Frequency Mapping Determined from 3D Synchrotron Radiation Phase-Contrast Imaging

OBJECTIVE

Cochlear implants are traditionally programmed to stimulate according to a generalized frequency map, where individual anatomic variability is not considered when selecting the centre frequency of stimulation of each implant electrode. However, high variability in cochlear size and spatial frequency distributions exist among individuals. Generalized cochlear implant frequency maps can result in large pitch perception errors and reduced hearing outcomes for cochlear implant recipients. The objective of this work was to develop an individualized frequency mapping technique for the human cochlea to allow for patient-specific cochlear implant stimulation.

METHODS

Ten cadaveric human cochleae were scanned using synchrotron radiation phase-contrast imaging (SR-PCI) combined with computed tomography (CT). For each cochlea, ground truth angle-frequency measurements were obtained in three-dimensions using the SR-PCI CT data. Using an approach designed to minimize perceptual error in frequency estimation, an individualized frequency function was determined to relate angular depth to frequency within the cochlea.

RESULTS

The individualized frequency mapping function significantly reduced pitch errors in comparison to the current gold standard generalized approach.

CONCLUSION AND SIGNIFICANCE

This paper presents for the first time a cochlear frequency map which can be individualized using only the angular length of cochleae. This approach can be applied in the clinical setting and has the potential to revolutionize cochlear implant programming for patients worldwide.

Intraoperative Correction of Cochlear Implant Electrode Translocation

INTRODUCTION

Translocation of precurved cochlear implant (CI) electrodes reduces hearing outcomes, but it is not known whether it is possible to correct scalar translocation such that all electrodes reside fully in the scala tympani (ST).

METHODS

Six cadaveric temporal bones were scanned with CT and segmented to delineate intracochlear anatomy. Mastoidectomy with facial recess was performed. Precurved CI electrodes (CI532; Cochlear Limited) were implanted until scalar translocation was confirmed with postoperative CT. Then, electrodes were removed and replaced. CT scan was repeated to assess for translocation correction. Scalar position of electrode contacts, angular insertion depth (AID) of the electrode array, and M- (average distance between each electrode contact and the modiolus) were measured. An in vivo case is reported in which intraoperative translocation detection led to removal and replacement of the electrode.

RESULTS

Five of 6 cadaveric translocations (83%) were corrected with 1 attempt, resulting in full ST insertions. AID averaged 285 ± 77° for translocated electrodes compared to 344 ± 28° for nontranslocated electrodes (p = 0.109). M- averaged 0.75 ± 0.18 mm for translocated electrodes and 0.45 ± 0.11 mm for nontranslocated electrodes (p = 0.016). Reduction in M- with translocation correction averaged 38%. In the in vivo case, translocation was successfully corrected in a single attempt.

CONCLUSION

Scalar translocation of precurved CI electrodes can be corrected by removal and reinsertion. This significantly improves the perimodiolar positioning of these electrodes. There was a high rate of success (83%) in this cadaveric model as well as a successful in vivo attempt.

On the Intracochlear Location of Straight Electrode Arrays After Cochlear Implantation: How Lateral Are Lateral Wall Electrodes?

OBJECTIVE

Cochlear implants are the gold standard for patients with severe sensorineural hearing loss. A focused electrical stimulation of individual spiral ganglion neurons has not been achieved yet because the scala tympani is a fluid-filled compartment and does not offer a matrix for neuritic outgrowth. Coating of the electrode contacts with swelling hydrogels could fill that gap between the electrode array and the medial wall of the cochlea. Therefore, the exact position of the electrode array within the scala tympani has to be known.

STUDY DESIGN

Retrospective analysis of patient data sets.

SETTING

Tertiary referral center. A total of 95 patients with cochlear implants from one manufacturer were included in this study. The lateral wall, the modiolar wall, and the cochlear implant electrode were segmented using OsiriX MD. For repositioning and reconstructing the respective contours and measuring distances, files were analyzed in MATLAB. The distances from the edge of each electrode contact to the cochlear walls showed no significant differences. But between the different contacts within each patient, there were significant differences. Around 180 degree insertion, electrodes start to get in contact with the lateral wall. The tip of the electrode array was always facing toward the modiolar wall independent of the length of the electrode. We established a method to analyze the position of electrodes within the cochlea.

Automatic segmentation of inner ear on CT-scan using auto-context convolutional neural network

Temporal bone CT-scan is a prerequisite in most surgical procedures concerning the ear such as cochlear implants. The 3D vision of inner ear structures is crucial for diagnostic and surgical preplanning purposes. Since clinical CT-scans are acquired at relatively low resolutions, improved performance can be achieved by registering patient-specific CT images to a high-resolution inner ear model built from accurate 3D segmentations based on micro-CT of human temporal bone specimens. This paper presents a framework based on convolutional neural network for human inner ear segmentation from micro-CT images which can be used to build such a model from an extensive database. The proposed approach employs an auto-context based cascaded 2D U-net architecture with 3D connected component refinement to segment the cochlear scalae, semicircular canals, and the vestibule. The system was formulated on a data set composed of 17 micro-CT from public Hear-EU dataset. A Dice coefficient of 0.90 and Hausdorff distance of 0.74 mm were obtained. The system yielded precise and fast automatic inner-ear segmentations.

Atlas-based segmentation of cochlear microstructures in cone beam CT

PURPOSE

To develop an automated segmentation approach for cochlear microstructures [scala tympani (ST), scala vestibuli (SV), modiolus (Mod), mid-modiolus (Mid-Mod), and round window membrane (RW)] in clinical cone beam computed tomography (CBCT) images of the temporal bone for use in surgical simulation software and for preoperative surgical evaluation.

METHODS

This approach was developed using the publicly available OpenEar (OE) Library that includes temporal bone specimens with spatially registered CBCT and 3D micro-slicing images. Five of these datasets were spatially aligned to our internal OSU atlas. An atlas of cochlear microstructures was created from one of the OE datasets. An affine registration of this atlas to the remaining OE CBCT images was used for automatically segmenting the cochlear microstructures. Quantitative metrics and visual review were used for validating the automatic segmentations.

RESULTS

The average DICE metrics were 0.77 and 0.74 for the ST and SV, respectively. The average Hausdorff distance (AVG HD) was 0.11 mm and 0.12 mm for both scalae. The mean distance between the centroids for the round window was 0.32 mm, and the mean AVG HD was 0.09 mm. The mean distance and angular rotation between the mid-modiolar axes were 0.11 mm and 9.8 degrees, respectively. Visually, the segmented structures were accurate and similar to that manually traced by an expert observer.

CONCLUSIONS

An atlas-based approach using 3D micro-slicing data and affine spatial registration in the cochlear region was successful in segmenting cochlear microstructures of temporal bone anatomy for use in simulation software and potentially for pre-surgical planning and rehearsal.

Practicable assessment of cochlear size and shape from clinical CT images

There is considerable interpersonal variation in the size and shape of the human cochlea, with evident consequences for cochlear implantation. The ability to characterize a specific cochlea, from preoperative computed tomography (CT) images, would allow the clinician to personalize the choice of electrode, surgical approach and postoperative programming. In this study, we present a fast, practicable and freely available method for estimating cochlear size and shape from clinical CT. The approach taken is to fit a template surface to the CT data, using either a statistical shape model or a locally affine deformation (LAD). After fitting, we measure cochlear size, duct length and a novel measure of basal turn non-planarity, which we suggest might correlate with the risk of insertion trauma. Gold-standard measurements from a convenience sample of 18 micro-CT scans are compared with the same quantities estimated from low-resolution, noisy, pseudo-clinical data synthesized from the same micro-CT scans. The best results were obtained using the LAD method, with an expected error of 8-17% of the gold-standard sample range for non-planarity, cochlear size and duct length.

IE-Map: a novel in-vivo atlas and template of the human inner ear

Brain atlases and templates are core tools in scientific research with increasing importance also in clinical applications. Advances in neuroimaging now allowed us to expand the atlas domain to the vestibular and auditory organ, the inner ear. In this study, we present IE-Map, an in-vivo template and atlas of the human labyrinth derived from multi-modal high-resolution magnetic resonance imaging (MRI) data, in a fully non-invasive manner without any contrast agent or radiation. We reconstructed a common template from 126 inner ears (63 normal subjects) and annotated it with 94 established landmarks and semi-automatic segmentations of all relevant macroscopic vestibular and auditory substructures. We validated the atlas by comparing MRI templates to a novel CT/micro-CT atlas, which we reconstructed from 21 publicly available post-mortem images of the bony labyrinth. Templates in MRI and micro-CT have a high overlap, and several key anatomical measures of the bony labyrinth in IE-Map are in line with micro-CT literature of the inner ear. A quantitative substructural analysis based on the new template, revealed a correlation of labyrinth parameters with total intracranial volume. No effects of gender or laterality were found. We provide the validated templates, atlas segmentations, surface meshes and landmark annotations as open-access material, to provide neuroscience researchers and clinicians in neurology, neurosurgery, and otorhinolaryngology with a widely applicable tool for computational neuro-otology.

Automatic segmentation of temporal bone structures from clinical conventional CT using a CNN approach

BACKGROUND

Automatic segmentation of temporal bone structures from patients' conventional computed tomography (CT) data plays an important role in the image-guided cochlear implant surgery. Existing convolutional neural network approaches have difficulties in segmenting such small tubular structures.

METHODS

We propose a light-weight three-dimensional convolutional neural network referred to as W-Net to achieve multiobjective segmentation of temporal bone structures including the cochlear labyrinth, ossicular chain and facial nerve from conventional temporal bone CT images. Data augmentation with morphological enhancement is proposed to increase the segmentation accuracy of small tubular structures. Evaluation against the state-of-the-art methods is performed.

RESULTS

Our method achieved mean Dice similarity coefficients (DSCs) of 0.90, 0.85 and 0.77 for the cochlear labyrinth, ossicular chain and facial nerve, respectively. These results were also close to the DSCs between human expert annotators (0.91, 0.91 and 0.72).

CONCLUSIONS

Our method achieves human-level accuracy in the segmentation of the cochlear labyrinth, ossicular chain and facial nerve.

Deep learning for the fully automated segmentation of the inner ear on MRI

Segmentation of anatomical structures is valuable in a variety of tasks, including 3D visualization, surgical planning, and quantitative image analysis. Manual segmentation is time-consuming and deals with intra and inter-observer variability. To develop a deep-learning approach for the fully automated segmentation of the inner ear in MRI, a 3D U-net was trained on 944 MRI scans with manually segmented inner ears as reference standard. The model was validated on an independent, multicentric dataset consisting of 177 MRI scans from three different centers. The model was also evaluated on a clinical validation set containing eight MRI scans with severe changes in the morphology of the labyrinth. The 3D U-net model showed precise Dice Similarity Coefficient scores (mean DSC-0.8790) with a high True Positive Rate (91.5%) and low False Discovery Rate and False Negative Rates (14.8% and 8.49% respectively) across images from three different centers. The model proved to perform well with a DSC of 0.8768 on the clinical validation dataset. The proposed auto-segmentation model is equivalent to human readers and is a reliable, consistent, and efficient method for inner ear segmentation, which can be used in a variety of clinical applications such as surgical planning and quantitative image analysis.

Fully automated preoperative segmentation of temporal bone structures from clinical CT scans

Middle- and inner-ear surgery is a vital treatment option in hearing loss, infections, and tumors of the lateral skull base. Segmentation of otologic structures from computed tomography (CT) has many potential applications for improving surgical planning but can be an arduous and time-consuming task. We propose an end-to-end solution for the automated segmentation of temporal bone CT using convolutional neural networks (CNN). Using 150 manually segmented CT scans, a comparison of 3 CNN models (AH-Net, U-Net, ResNet) was conducted to compare Dice coefficient, Hausdorff distance, and speed of segmentation of the inner ear, ossicles, facial nerve and sigmoid sinus. Using AH-Net, the Dice coefficient was 0.91 for the inner ear; 0.85 for the ossicles; 0.75 for the facial nerve; and 0.86 for the sigmoid sinus. The average Hausdorff distance was 0.25, 0.21, 0.24 and 0.45 mm, respectively. Blinded experts assessed the accuracy of both techniques, and there was no statistical difference between the ratings for the two methods (p = 0.93). Objective and subjective assessment confirm good correlation between automated segmentation of otologic structures and manual segmentation performed by a specialist. This end-to-end automated segmentation pipeline can help to advance the systematic application of augmented reality, simulation, and automation in otologic procedures.

A Graph-Based Method for Optimal Active Electrode Selection in Cochlear Implants

The cochlear implant (CI) is a neural prosthetic that is the standard-of-care treatment for severe-to-profound hearing loss. CIs consist of an electrode array inserted into the cochlea that electrically stimulates auditory nerve fibers to induce the sensation of hearing. Competing stimuli occur when multiple electrodes stimulate the same neural pathways. This is known to negatively impact hearing outcomes. Previous research has shown that image-processing techniques can be used to analyze the CI position in CT scans to estimate the degree of competition between electrodes based on the CI user's unique anatomy and electrode placement. The resulting data permits an algorithm or expert to select a subset of electrodes to keep active to alleviate competition. Expert selection of electrodes using this data has been shown in clinical studies to lead to significantly improved hearing outcomes for CI users. Currently, we aim to translate these techniques to a system designed for worldwide clinical use, which mandates that the selection of active electrodes be automated by robust algorithms. Previously proposed techniques produce optimal plans with only 48% success rate. In this work, we propose a new graph-based approach. We design a graph with nodes that represent electrodes and edge weights that encode competition between electrode pairs. We then find an optimal path through this graph to determine the active electrode set. Our method produces results judged by an expert to be optimal in over 95% of cases. This technique could facilitate widespread clinical translation of image-guided cochlear implant programming methods.

Hearing Preservation Outcomes Using a Precurved Electrode Array Inserted With an External Sheath

OBJECTIVES

Describe audiologic outcomes in hearing preservation cochlear implantation (CI) using a precurved electrode array inserted using an external sheath and evaluate association of electrode positioning and preservation of residual hearing.

STUDY DESIGN

Retrospective review.

SETTING

Tertiary otologic center.

PATIENTS

Twenty-four adult patients who underwent hearing preservation CI with precurved electrode array.

INTERVENTIONS

CI, intraoperative computed tomography (CT) OUTCOME MEASURES:: Audiologic measures (consonant-nucleus-consonant [CNC] words, AzBio sentences, low-frequency pure tone averages [LFPTA]) and electrode location (scalar location, electrode-to-modiolus distance ((Equation is included in full-text article.)), angular insertion depth).

RESULTS

Twenty-four adults with less than 80 dB LFPTA with a precurved electrode array inserted using an external sheath; 16 underwent intraoperative CT. LFPTA was 58.5 dB HL preoperatively, with a 17.3 dB threshold shift at CI activation (p = 0.005). CNC word scores improved from 6% preoperatively to 64% at 6 months postoperatively (p < 0.0001). There was one scalar translocation and no tip fold-overs. The average angular insertion depth was 388.2 degrees, and the average (Equation is included in full-text article.)across all electrodes was 0.36 mm. Multivariate regression revealed a significant correlation between CNC scores at 6 months and angular insertion depth (p = 0.0122; r = 0.45, adjusted r = 0.35). Change in LFPTA was not significantly associated with angular insertion depth or (Equation is included in full-text article.).

CONCLUSIONS

A low rate of translocation allows a precurved electrode array inserted using an external sheath to maintain hearing preservation rates comparable to straight electrode arrays. With scala tympani insertion, angular insertion depth is a positive marker of improved speech performance postoperatively but may be a confounder variable based on individual cochlear size.

Multi-Scale deep learning framework for cochlea localization, segmentation and analysis on clinical ultra-high-resolution CT images

BACKGROUND AND OBJECTIVE

Performing patient-specific, pre-operative cochlea CT-based measurements could be helpful to positively affect the outcome of cochlear surgery in terms of intracochlear trauma and loss of residual hearing. Therefore, we propose a method to automatically segment and measure the human cochlea in clinical ultra-high-resolution (UHR) CT images, and investigate differences in cochlea size for personalized implant planning.

METHODS

123 temporal bone CT scans were acquired with two UHR-CT scanners, and used to develop and validate a deep learning-based system for automated cochlea segmentation and measurement. The segmentation algorithm is composed of two major steps (detection and pixel-wise classification) in cascade, and aims at combining the results of a multi-scale computer-aided detection scheme with a U-Net-like architecture for pixelwise classification. The segmentation results were used as an input to the measurement algorithm, which provides automatic cochlear measurements (volume, basal diameter, and cochlear duct length (CDL)) through the combined use of convolutional neural networks and thinning algorithms. Automatic segmentation was validated against manual annotation, by the means of Dice similarity, Boundary-F1 (BF) score, and maximum and average Hausdorff distances, while measurement errors were calculated between the automatic results and the corresponding manually obtained ground truth on a per-patient basis. Finally, the developed system was used to investigate the differences in cochlea size within our patient cohort, to relate the measurement errors to the actual variation in cochlear size across different patients.

RESULTS

Automatic segmentation resulted in a Dice of 0.90 ± 0.03, BF score of 0.95 ± 0.03, and maximum and average Hausdorff distance of 3.05 ± 0.39 and 0.32 ± 0.07 against manual annotation. Automatic cochlear measurements resulted in errors of 8.4% (volume), 5.5% (CDL), 7.8% (basal diameter). The cochlea size varied broadly, ranging between 0.10 and 0.28 ml (volume), 1.3 and 2.5 mm (basal diameter), and 27.7 and 40.1 mm (CDL).

CONCLUSIONS

The proposed algorithm could successfully segment and analyze the cochlea on UHR-CT images, resulting in accurate measurements of cochlear anatomy. Given the wide variation in cochlear size found in our patient cohort, it may find application as a pre-operative tool in cochlear implant surgery, potentially helping elaborate personalized treatment strategies based on patient-specific, image-based anatomical measurements.

Insertion Depth for Optimized Positioning of Precurved Cochlear Implant Electrodes

HYPOTHESIS

Generic guidelines for insertion depth of precurved electrodes are suboptimal for many individuals.

BACKGROUND

Insertion depths that are too shallow result in decreased cochlear coverage, and ones that are too deep lift electrodes away from the modiolus and degrade the electro-neural interface. Guidelines for insertion depth are generically applied to all individuals using insertion depth markers on the array that can be referenced against anatomical landmarks.

METHODS

To normalize our measurements, we determined the optimal position and insertion vector where a precurved array best fits the cochlea for each patient in an IRB-approved, N = 131 subject CT database. The distances from the most basal electrode on an optimally placed array to anatomical landmarks, including the round window (RW) and facial recess (FR), was measured for all patients.

RESULTS

The standard deviations of the distance from the most basal electrode to the FR and RW are 0.65 mm and 0.26 mm, respectively. Owing to the high variability in FR distance, using the FR as a landmark to determine insertion depth results in >0.5 mm difference with ideal depth in 44% of cases. Alignment of either of the two most proximal RW markers with the RW would result in over-insertion failures for >80% of cases, whereas the use of the third, most medial marker would result in under-insertion in only 19% of cases.

CONCLUSIONS

Normalized measurements using the optimized insertion vector show low variance in distance from the basal electrode position to the RW, thereby suggesting it as a better landmark for determining insertion depth than the FR.

Preoperative prediction of angular insertion depth of lateral wall cochlear implant electrode arrays

Purpose: Cochlear implants (CIs) use an array of electrodes surgically threaded into the cochlea to restore hearing sensation. Techniques for predicting the insertion depth of the array into the cochlea could guide surgeons toward more optimal placement of the array to reduce trauma and preserve the residual hearing. In addition to the electrode array geometry, the base insertion depth (BID) and the cochlear size could impact the overall array insertion depth. Approach: We investigated using these measurements to develop a linear regression model that can make preoperative or intraoperative predictions of the insertion depth of lateral wall CI electrodes. Computed tomography (CT) images of 86 CI recipients were analyzed. Using previously developed automated algorithms, the relative electrode position inside the cochlea was measured from the CT images. Results: A linear regression model is proposed for insertion depth prediction based on cochlea size, array geometry, and BID. The model is able to accurately predict angular insertion depths with a standard deviation of 41 deg and absolute deviation error of 32 deg. Conclusions: Surgeons may use this model for patient-customized selection of electrode array and/or to plan a BID for a given array that minimizes the likelihood of causing trauma to regions of the cochlea where residual hearing exists.

HeadLocNet: Deep convolutional neural networks for accurate classification and multi-landmark localization of head CTs

Cochlear implants (CIs) are used to treat subjects with hearing loss. In a CI surgery, an electrode array is inserted into the cochlea to stimulate auditory nerves. After surgery, CIs need to be programmed. Studies have shown that the cochlea-electrode spatial relationship derived from medical images can guide CI programming and lead to significant improvement in hearing outcomes. We have developed a series of algorithms to segment the inner ear anatomy and localize the electrodes. But, because clinical head CT images are acquired with different protocols, the field of view and orientation of the image volumes vary greatly. As a consequence, visual inspection and manual image registration to an atlas image are needed to document their content and to initialize intensity-based registration algorithms used in our processing pipeline. For large-scale evaluation and deployment of our methods these steps need to be automated. In this article we propose to achieve this with a deep convolutional neural network (CNN) that can be trained end-to-end to classify a head CT image in terms of its content and to localize landmarks. The detected landmarks can then be used to estimate a point-based registration with the atlas image in which the same landmark set's positions are known. We achieve 99.5% classification accuracy and an average localization error of 3.45 mm for 7 landmarks located around each inner ear. This is better than what was achieved with earlier methods we have proposed for the same tasks.

Use of intraoperative CT scanning for quality control assessment of cochlear implant electrode array placement

Background: Imaging of cochlear implant (CI) electrode arrays (EAs) consists of intraoperative fluoroscopy to rule out tip fold-over and/or post-operative computerized tomography (CT) if concern exists regarding extrusion or misplacement of the EA. Intraoperative CT (iCT) can satisfy these current needs and enables specification of final intracochlear position.Aims/objectives: To describe iCT scanning of CI recipients at an academic medical center.Materials and methods: iCT was used to scan CI recipients within the operating room before recovering from general anesthesia.Results: In fiscal year 2019, 301 CI were placed (83 children, 218 adult). One hundred, seventy-five iCTs were performed (58% of total CIs) of which 52 were children (63% of pediatric CIs) and 123 were adult (57% of adult CIs). Of 7 CI surgeons, use of iCT ranged from 14% to 100% (mean 60%). Four tip fold-overs were identified and corrected intraoperatively. Surgeons reported using the images to improve technique (i.e. pulling back precurved EAs to improve perimodiolar positioning).Conclusion and significance: The current standard of care for CI is to insert EAs without feedback as to final location. iCT provides surgeons with rapid post-insertion feedback which allows detection and correction of suboptimally placed EAs as well as refinement of surgical technique.

Effect of Scala Tympani Height on Insertion Depth of Straight Cochlear Implant Electrodes

OBJECTIVE

Studies suggest lateral wall (LW) scala tympani (ST) height decreases apically, which may limit insertion depth. No studies have investigated the relationship of LW ST height with translocation rate or location.

STUDY DESIGN

Retrospective review.

SETTING

Cochlear implant program at tertiary referral center.

SUBJECTS AND METHODS

LW ST height was measured in preoperative images for patients with straight electrodes. Scalar location, angle of insertion depth (AID), and translocation depth were measured in postoperative images. Audiologic outcomes were tracked.

RESULTS

In total, 177 ears were identified with 39 translocations (22%). Median AID was 443° (interquartile range [IQR], 367°-550°). Audiologic outcomes (126 ears) showed a small, significant correlation between consonant-nucleus-consonant (CNC) word score and AID (r = 0.20, P = .027), although correlation was insignificant if translocation occurred (r = 0.11, P = .553). Translocation did not affect CNC score (P = .335). AID was higher for translocated electrodes (503° vs 445°, P = .004). Median translocation depth was 381° (IQR, 222°-399°). Median depth at which a 0.5-mm electrode would not fit within 0.1 mm of LW was 585° (IQR, 405°-585°). Median depth at which a 0.5-mm electrode would displace the basilar membrane by ≥0.1 mm was 585° (IQR, 518°-765°); this was defined as predicted translocation depth (PTD). Translocation rate was 39% for insertions deeper than PTD and 14% for insertions shallower than PTD (P = .008).

CONCLUSION

AID and CNC are directly correlated for straight electrodes when not translocated. Translocations generally occur around 380° and are more common with deeper insertions due to decreasing LW ST height. Risk of translocation increases significantly after 580°.

Magnetically Steered Robotic Insertion of Cochlear-Implant Electrode Arrays: System Integration and First-In-Cadaver Results

Cochlear-implant electrode arrays (EAs) must be inserted accurately and precisely to avoid damaging the delicate anatomical structures of the inner ear. It has previously been shown on the benchtop that using magnetic fields to steer magnet-tipped EAs during insertion reduces insertion forces, which correlate with insertion errors and damage to internal cochlear structures. This paper presents several advancements toward the goal of deploying magnetic steering of cochlear-implant EAs in the operating room. In particular, we integrate image guidance with patient-specific insertion vectors, we incorporate a new nonmagnetic insertion tool, and we use an electromagnetic source, which provides programmable control over the generated field. The electromagnet is safer than prior permanent-magnet approaches in two ways: it eliminates motion of the field source relative to the patient's head and creates a field-free source in the power-off state. Using this system, we demonstrate system feasibility by magnetically steering EAs into a cadaver cochlea for the first time. We show that magnetic steering decreases average insertion forces, in comparison to manual insertions and to image-guided robotic insertions alone.

A 3D deep supervised densely network for small organs of human temporal bone segmentation in CT images

Computed Tomography (CT) has become an important way for examining the critical anatomical organs of the human temporal bone in the diagnosis and treatment of ear diseases. Segmentation of the critical anatomical organs is an important fundamental step for the computer assistant analysis of human temporal bone CT images. However, it is challenging to segment sophisticated and small organs. To deal with this issue, a novel 3D Deep Supervised Densely Network (3D-DSD Net) is proposed in this paper. The network adopts a dense connection design and a 3D multi-pooling feature fusion strategy in the encoding stage of the 3D-Unet, and a 3D deep supervised mechanism is employed in the decoding stage. The experimental results show that our method achieved competitive performance in the CT data segmentation task of the small organs in the temporal bone.

Metal artifact reduction for the segmentation of the intra cochlear anatomy in CT images of the ear with 3D-conditional GANs

Cochlear implants (CIs) are surgically implanted neural prosthetic devices that are used to treat severe-to-profound hearing loss. These devices are programmed post implantation and precise knowledge of the implant position with respect to the intra cochlear anatomy (ICA) can help the programming audiologists. Over the years, we have developed algorithms that permit determining the position of implanted electrodes relative to the ICA using pre- and post-implantation CT image pairs. However, these do not extend to CI recipients for whom pre-implantation CT (Pre-CT) images are not available. This is so because post-operative images are affected by strong artifacts introduced by the metallic implant. To overcome this issue, we have proposed two methods to segment the ICA in post-implantation CT (Post-CT) images, but they lead to segmentation errors that are substantially larger than errors obtained with Pre-CT images. Recently, we have proposed an approach that uses 2D-conditional generative adversarial nets (cGANs) to synthesize pre-operative images from post-operative images. This permits to use segmentation algorithms designed to operate on Pre-CT images even when these are not available. We have shown that it substantially and significantly improves the results obtained with methods designed to operate directly on post-CT images. In this article, we expand on our earlier work by moving from a 2D architecture to a 3D architecture. We perform a large validation and comparative study that shows that the 3D architecture improves significantly the quality of the synthetic images measured by the commonly used MSSIM (Mean Structural SIMilarity index). We also show that the segmentation results obtained with the 3D architecture are better than those obtained with the 2D architecture although differences have not reached statistical significance.

Preliminary Results With Image-guided Cochlear Implant Insertion Techniques

HYPOTHESIS

Using patient-customized cochlear measurements obtained from preoperative computed tomography (CT) scans to guide insertion of cochlear implant (CI) electrode arrays will lead to more optimal intracochlear positioning.

BACKGROUND

Cochlear duct length is highly variable ranging from 25.26 to 35.46 mm, yet CI electrode arrays are treated as one size fits most. We sought to investigate the impact of patient-customized insertion plans on final location of electrode arrays.

METHODS

Twenty cadaveric temporal bone specimens were CT scanned and randomly divided into groups A and B. Group A specimens had an optimal customized insertion plan generated including entry site (e.g., round window versus extended round window), entry vector based on anatomical landmarks (e.g., hug posterior aspect of facial recess and angle 1 mm inferior to stapes), depth to begin advancing off stylet, and final insertion depth. Suboptimal plans were chosen for group B by selecting an approach that was normal yet predicted to result in poor final electrode location. One surgeon, blinded as to group, carried out the CI insertions following which the electrode array was fixed using superglue and the specimen CT scanned to allow assessment of final electrode location.

RESULTS

Average perimodiolar distances for groups A and B were 0.51 and 0.60 mm, respectively. For group A, full scala tympani insertion was achieved in all specimens while in group B, 4 of 10 specimens had scalar translocation.

CONCLUSION

Patient customized cochlear implant insertion techniques achieved better positioning of electrode arrays in this study and have potential for improving electrode positioning in patients.

Prevalence of Extracochlear Electrodes: Computerized Tomography Scans, Cochlear Implant Maps, and Operative Reports

OBJECTIVE

To quantify and compare the number of cochlear implant (CI) electrodes found to be extracochlear on postoperative computerized tomography (CT) scans, the number of basal electrodes deactivated during standard CI mapping (without knowledge of the postoperative CT scan), and the extent of electrode insertion noted by the surgeon.

STUDY DESIGN

Retrospective.

SETTING

Academic Medical Center.

METHODS

Two hundred sixty-two patients underwent standard cochlear implantation and postoperative temporal bone CT scanning. Scans were analyzed to determine the number of extracochlear electrodes. Standard CI programming had been completed without knowledge of the extracochlear electrodes identified on the CT. These standard CI maps were reviewed to record the number of deactivated basal electrodes. Lastly, each operative report was reviewed to record the extent of reported electrode insertion.

RESULTS

13.4% (n = 35) of CIs were found to have at least one electrode outside of the cochlea on the CT scan. Review of CI mapping indicated that audiologists had deactivated extracochlear electrodes in 60% (21) of these cases. Review of operative reports revealed that surgeons correctly indicated the number of extracochlear electrodes in 6% (2) of these cases.

CONCLUSIONS

Extracochlear electrodes were correctly identified audiologically in 60% of cases and in surgical reports in 6% of cases; however, it is possible that at least a portion of these cases involved postoperative electrode migration. Given these findings, postoperative CT scans can provide information regarding basal electrode location, which could help improve programming accuracy, associated frequency allocation, and audibility with appropriate deactivation of extracochlear electrodes.

An In-Vitro Insertion-Force Study of Magnetically Guided Lateral-Wall Cochlear-Implant Electrode Arrays

Hypothesis:

Insertion forces can be reduced by magnetically guiding the tip of lateral-wall cochlear-implant electrode arrays during insertion via both cochleostomy and the round window.

Background:

Steerable electrode arrays have the potential to minimize intracochlear trauma by reducing the severity of contact between the electrode-array tip and the cochlear wall. However, steerable electrode arrays typically have increased stiffness associated with the steering mechanism. In addition, steerable electrode arrays are typically designed to curve in the direction of the basal turn, which is not ideal for round-window insertions, as the cochlear hook's curvature is in the opposite direction. Lateral-wall electrode arrays can be modified to include magnets at their tips, augmenting their superior flexibility with a steering mechanism. By applying magnetic torque to the tip, an electrode array can be navigated through the cochlear hook and the basal turn.

Methods:

Automated insertions of candidate electrode arrays are conducted into a scala-tympani phantom with either a cochleostomy or round-window opening. The phantom is mounted on a multi-degree-of-freedom force sensor. An external magnet applies the necessary magnetic bending torque to the magnetic tip of a modified clinical electrode array, coordinated with the insertion, with the goal of directing the tip down the lumen. Steering of the electrode array is verified through a camera.

Results:

Statistical t-test results indicate that magnetic guidance does reduce insertion forces by as much as 50% with certain electrode-array models. Direct tip contact with the medial wall through the cochlear hook and the lateral wall of the basal turn is completely eliminated. The magnetic field required to accomplish these insertions varied from 77 to 225 mT based on the volume of the magnet at the tip of the electrode array. Alteration of the tip to accommodate a tiny magnet is minimal and does not change the insertion characteristic of the electrode array unless the tip shape is altered.

Conclusion:

Magnetic guidance can eliminate direct tip contact with the medial walls through the cochlear hook and the lateral walls of the basal turn. Insertion-force reduction will vary based on the electrode-array model, but is statistically significant for all models tested. Successful steering of lateral-wall electrode arrays is accomplished while maintaining its superior flexibility.

Further Evidence of the Relationship Between Cochlear Implant Electrode Positioning and Hearing Outcomes

BACKGROUND

Postoperative imaging studies by numerous groups have revealed that final cochlear implant (CI) electrode position impacts audiological outcomes with scalar location consistently shown to be a significant factor. Modiolar proximity has been less extensively studied, and findings regarding the effect of insertion depth have been inconsistent.

METHODS

Using previously developed automated algorithms, we determined CI electrode position in an Institutional Review Board-approved database of 220 CI ears. Generalized linear models (GLM) were used to analyze the relationship between audiological outcomes and factors including age, duration of CI use, device type, and electrode position.

RESULTS

For precurved arrays, GLM revealed that scalar position, modiolar proximity, base insertion depth, and sex were significant factors for Consonant-Nucleus-Consonant (CNC) words (R = 0.43, p < 0.001, n = 92 arrays), while scalar position, modiolar proximity, age, and postlingual onset of deafness were significant for Bamford-Kawal-Bench Sentences in Noise (BKB-SIN) (R = 0.51, p < 0.001, n = 85) scores. Other factors were not significant in the final model after controlling for these variables. For straight arrays, we found the insertion depth, postlingual deafness, and length of CI use to be highly significant (R = 0.47, p < 0.001) factors for CNC words (91 arrays), while for BKB-SIN scores the most significant (R = 0.47, p < 0.001) factors were insertion depth, younger age, and postlingual deafness (89 arrays).

CONCLUSION

Our results confirm the significance of electrode positioning in audiological outcomes. The most significant positional predictors of outcome for precurved arrays were full scala tympani (ST) insertion and the modiolar distance, while for the lateral wall arrays the depth of insertion was the most significant factor.