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

Neves, Caio Athayde; Tran, Emma; Kessler, Iruena Moraes; Blevins, Nikolas H.

doi:10.1038/s41598-020-80619-0

Cited by 44 publications

(90 citation statements)

References 35 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Temporal bone thickness has been extensively studied in the context of otological surgery (5,34,(36)(37)(38). Our study reproduces the known variability, showing the largest available bone thickness within a radius of 19 mm from Henle's Spine.…”

Section: Temporal Bone Thicknesssupporting

confidence: 70%

See 1 more Smart Citation

Quantitative Analysis of Temporal Bone Density and Thickness for Robotic Ear Surgery

et al. 2021

View full text Add to dashboard Cite

Background and Objective: Quantitative assessment of bone density and thickness in computed-tomography images offers great potential for preoperative planning procedures in robotic ear surgery.Methods: We retrospectively analyzed computed-tomography scans of subjects undergoing cochlear implantation (N = 39). In addition, scans of Thiel-fixated ex-vivo specimens were analyzed (N = 15). To estimate bone mineral density, quantitative computed-tomography data were obtained using a calibration phantom. The temporal bone thickness and cortical bone density were systematically assessed at retroauricular positions using an automated algorithm referenced by an anatomy-based coordinate system. Two indices are proposed to include information of bone density and thickness for the preoperative assessment of safe screw positions (Screw Implantation Safety Index, SISI) and mass distribution (Column Density Index, CODI). Linear mixed-effects models were used to assess the effects of age, gender, ear side and position on bone thickness, cortical bone density and the distribution of the indices.Results: Age, gender, and ear side only had negligible effects on temporal bone thickness and cortical bone density. The average radiodensity of cortical bone was 1,511 Hounsfield units, corresponding to a bone mineral density of 1,145 mg HA/cm3. Temporal bone thickness and cortical bone density depend on the distance from Henle's spine in posterior direction. Moreover, safe screw placement locations can be identified by computation of the SISI distribution. A local maximum in mass distribution was observed posteriorly to the supramastoid crest.Conclusions: We provide quantitative information about temporal bone density and thickness for applications in robotic and computer-assisted ear surgery. The proposed preoperative indices (SISI and CODI) can be applied to patient-specific cases to identify optimal regions with respect to bone density and thickness for safe screw placement and effective implant positioning.

show abstract

Section: Temporal Bone Thicknesssupporting

confidence: 70%

“…Other transfer methods, such as template-guided approaches (33), could also be used. For patient-specific planning, the proposed methods and indices could be computed using automated segmentation tools (34,35).…”

Section: Discussionmentioning

confidence: 99%

Quantitative Analysis of Temporal Bone Density and Thickness for Robotic Ear Surgery

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The DL group (n = 165) was used to generate an automated prediction model of the cochlea and the otic capsule as previously described by our group. 27 This DL model was used to segment the cochlea and the otic capsule of the Testing group, and the preoperative CTs from the Validation group. The 3D anatomic representations of the cochlea and otic capsule were exported as polygonal models for geometric analysis in OBJ file format.…”

Section: Segmentation and Model Generationmentioning

confidence: 99%

Fully Automated Measurement of Cochlear Duct Length From Clinical Temporal Bone Computed Tomography

et al. 2021

View full text Add to dashboard Cite

Objectives/Hypothesis: To present and validate a novel fully automated method to measure cochlear dimensions, including cochlear duct length (CDL).Study Design: Cross-sectional study.Methods: The computational method combined 1) a deep learning (DL) algorithm to segment the cochlea and otic capsule and 2) geometric analysis to measure anti-modiolar distances from the round window to the apex. The algorithm was trained using 165 manually segmented clinical computed tomography (CT). A Testing group of 159 CTs were then measured for cochlear diameter and width (A-and B-values) and CDL using the automated system and compared against manual measurements. The results were also compared with existing approaches and historical data. In addition, pre-and postimplantation scans from 27 cochlear implant recipients were studied to compare predicted versus actual array insertion depth.Results: Measurements were successfully obtained in 98.1% of scans. The mean CDL to 900 was 35.52 mm (SD, 2.06; range, [30.91-40.50]), the mean A-value was 8.88 mm (0.47; [7.67-10.49]), and mean B-value was 6.38 mm (0.42; [5.16-7.38]). The R 2 fit of the automated to manual measurements was 0.87 for A-value, 0.70 for B-value, and 0.71 for CDL. For anti-modiolar arrays, the distance between the imaged and predicted array tip location was 0.57 mm (1.25; [0.13-5.28]). Conclusion:Our method provides a fully automated means of cochlear analysis from clinical CTs. The distribution of CDL, dimensions, and cochlear quadrant lengths is similar to those from historical data. This approach requires no radiographic experience and is free from user-related variation.

show abstract

“…We also compare the performance of our proposed network and the other state-of-the-art CNN architectures. The semicircular canal can be segmented effectively with the proposed method, so can the other organs in the future, for example, facial nerve [ 27 ], cochleae [ 28 ], and spinal cord [ 29 ]. …”

Section: Introductionmentioning

confidence: 99%

“…The semicircular canal can be segmented effectively with the proposed method, so can the other organs in the future, for example, facial nerve [ 27 ], cochleae [ 28 ], and spinal cord [ 29 ].…”

Section: Introductionmentioning

confidence: 99%

Automatic Semicircular Canal Segmentation of CT Volumes Using Improved 3D U‐Net with Attention Mechanism

Liu

Chen

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The vestibular system is the sensory apparatus that helps the body maintain its postural equilibrium, and semicircular canal is an important organ of the vestibular system. The semicircular canals are three membranous tubes, each forming approximately two-thirds of a circle with a diameter of approximately 6.5 mm, and segmenting them accurately is of great benefit for auxiliary diagnosis, surgery, and treatment of vestibular disease. However, the semicircular canal has small volume, which accounts for less than 1% of the overall computed tomography image. Doctors have to annotate the image in a slice-by-slice manner, which is time-consuming and labor-intensive. To solve this problem, we propose a novel 3D convolutional neural network based on 3D U-Net to automatically segment the semicircular canal. We added the spatial attention mechanism of 3D spatial squeeze and excitation modules, as well as channel attention mechanism of 3D global attention upsample modules to improve the network performance. Our network achieved an average dice coefficient of 92.5% on the test dataset, which shows competitive performance in semicircular canals segmentation task.

show abstract

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

Cited by 44 publications

References 35 publications

Quantitative Analysis of Temporal Bone Density and Thickness for Robotic Ear Surgery

Quantitative Analysis of Temporal Bone Density and Thickness for Robotic Ear Surgery

Fully Automated Measurement of Cochlear Duct Length From Clinical Temporal Bone Computed Tomography

Automatic Semicircular Canal Segmentation of CT Volumes Using Improved 3D U‐Net with Attention Mechanism

Contact Info

Product

Resources

About