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

Abstract 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.

Download Full-text

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

Scientific Reports ◽

10.1038/s41598-020-80619-0 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

C. A. Neves ◽

E. D. Tran ◽

I. M. Kessler ◽

N. H. Blevins

Keyword(s):

Facial Nerve ◽

Inner Ear ◽

Temporal Bone ◽

Hausdorff Distance ◽

Sigmoid Sinus ◽

Ct Scans ◽

Automated Segmentation ◽

Dice Coefficient ◽

Lateral Skull Base ◽

End To End

AbstractMiddle- 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.

Download Full-text

Congenital unilateral facial palsy revealing a facial nerve agenesis: a case report and review of the literature

BJR|case reports ◽

10.1259/bjrcr.20180029 ◽

2019 ◽

Vol 5 (1) ◽

pp. 20180029

Author(s):

Yaotse Elikplim Nordjoe ◽

Ouidad Azdad ◽

Mohamed Lahkim ◽

Laila Jroundi ◽

Fatima Zahrae Laamrani

Keyword(s):

Case Report ◽

Facial Nerve ◽

Temporal Bone ◽

Facial Paralysis ◽

Facial Palsy ◽

Rare Condition ◽

Review Of The Literature ◽

Moebius Syndrome ◽

Temporal Bone Ct

Facial nerve aplasia is an extremely rare condition that is usually syndromic, namely, in Moebius syndrome. The occurrence of isolated agenesis of facial nerve is even rarer, with only few cases reported in the literature. We report a case of congenital facial paralysis due to facial nerve aplasia diagnosed on MRI, while no noticeable abnormality was detected on the temporal bone CT.

Download Full-text

The depth of the facial nerve in the mastoid bone

International Journal of Otorhinolaryngology and Head and Neck Surgery ◽

10.18203/issn.2454-5929.ijohns20181009 ◽

2018 ◽

Vol 4 (3) ◽

pp. 666

Author(s):

B. Y. Praveen Kumar ◽

K. T. Chandrashekhar ◽

M. K. Veena Pani ◽

Sunil K. C. ◽

Anand Kumar S. ◽

...

Keyword(s):

Facial Nerve ◽

Temporal Bone ◽

Significant Variation ◽

Canal Wall ◽

Stylomastoid Foramen ◽

Cortical Mastoidectomy ◽

Canal Wall Down ◽

Canal Wall Down Mastoidectomy ◽

The Mean ◽

Temporal Bones

Background: The hallmark of the temporal bone is variation. Various important structures like the facial nerve run in the temporal bone at various depths which can be injured during mastoidectomy.Methods: Twenty wet cadaveric temporal bones were dissected. A cortical mastoidectomy was performed followed by a canal wall down mastoidectomy and the depth of the vertical segment of the facial nerve in the mastoid was determined. Results: The mean depth of the second genu was 13.82 mm. The mean depth of the stylomastoid foramen was 12.75 mm and the mean distance from the annulus at 6’0 clock to the stylomastoid foramen was 10.22 mm.Conclusions: There is significant variation in the average depth of the facial nerve in the mastoid.

Download Full-text

Unilateral internal auditory canal stenosis with progressive facial weakness

Bangladesh Journal of Otorhinolaryngology ◽

10.3329/bjo.v18i2.12014 ◽

2012 ◽

Vol 18 (2) ◽

pp. 179-182

Author(s):

Sathiya Murali ◽

Arpana Shekhar ◽

S Shyam Sudhakar ◽

Kiran Natarajan ◽

Mohan Kameswaran

Keyword(s):

Hearing Loss ◽

High Resolution ◽

Facial Nerve ◽

Temporal Bone ◽

Internal Auditory Canal ◽

Facial Weakness ◽

Profound Hearing Loss ◽

Canal Stenosis ◽

Temporal Bone Ct ◽

Magnetic Resonance Imaging Mri

Internal auditory canal (IAC) stenosis is a rare cause of sensorineural hearing loss. Patient may present with symptoms of progressive facial nerve palsy, hearing loss, tinnitus and giddiness. High resolution temporal bone CT-scan and magnetic resonance imaging (MRI) are the important tools for diagnosis. No specific management has been devised. Here is presentation of a case of unilateral (left) IAC stenosis with profound hearing loss and progressive House Brackmann Grade III-IV facial weakness. The diameter of the IAC was less than 2 mm on high resolution temporal bone computed tomography (HRCT) scan. It was managed by facial nerve decompression by translabyrinthine approach in an attempt to prevent further deterioration of facial palsy. DOI: http://dx.doi.org/10.3329/bjo.v18i2.12014 Bangladesh J Otorhinolaryngol 2012; 18(2): 179-182

Download Full-text

Surgeon Oriented Preoperative Radiologic Evaluation in Cochlear Implantation - Our experience with a Proposed Checklist

International Archives of Otorhinolaryngology ◽

10.1055/s-0038-1648247 ◽

2018 ◽

Vol 23 (02) ◽

pp. 137-141 ◽

Cited By ~ 1

Author(s):

Mahmoud Mandour ◽

Mohammed Tomoum ◽

Saad El Zayat ◽

Hisham Hamad ◽

Mohamed Amer

Keyword(s):

Facial Nerve ◽

Temporal Bone ◽

Cochlear Implantation ◽

Round Window ◽

Special Focus ◽

Radiologic Evaluation ◽

Cortical Mastoidectomy ◽

Temporal Bone Ct ◽

Operative Findings ◽

Key Steps

Introduction Preoperative temporal bone imaging studies have been routinely performed prior to cochlear implantation. Radiologists need to report these examinations with special focus on the surgeon's expectations. Objectives To provide a basic structured format, in the form of a checklist, for reporting preoperative computed tomography (CT) and to its clinical impact on operative findings. Methods The preoperative temporal bone CT scans of 47 patients were analyzed and reported according to the proposed checklist. Intraoperative assessment of mastoidectomy, posterior tympanotomy and round window access was done by the surgeon in a blinded fashion and were correlated with the radiological findings to assess its significance. Results The proposed radiological checklist was reliable in assessing operative difficulty during cochlear implantation. Contracted mastoid and lower tegmen position were associated with a greater difficulty of the cortical mastoidectomy. Presence of an air cell around the facial nerve (FN) was predictive of easier facial recess access exposure. Facial nerve location and posterior external auditory canal (EAC) wall inclination were predictive of difficult round window (RW) accessibility. Conclusion Certain parameters on the preoperative temporal bone CT scan may be useful in predicting potential difficulties encountered during the key steps involved in cochlear implant surgery.

Download Full-text

Automated Registration-Based Temporal Bone Computed Tomography Segmentation for Applications in Neurotologic Surgery

Otolaryngology ◽

10.1177/01945998211044982 ◽

2021 ◽

pp. 019459982110449

Author(s):

Andy S. Ding ◽

Alexander Lu ◽

Zhaoshuo Li ◽

Deepa Galaiya ◽

Jeffrey H. Siewerdsen ◽

...

Keyword(s):

Computed Tomography ◽

Facial Nerve ◽

Temporal Bone ◽

Ground Truth ◽

Ct Images ◽

Cone Beam ◽

Data Set ◽

Registration Method ◽

Temporal Bone Ct ◽

Temporal Bone Anatomy

Objective This study investigates the accuracy of an automated method to rapidly segment relevant temporal bone anatomy from cone beam computed tomography (CT) images. Implementation of this segmentation pipeline has potential to improve surgical safety and decrease operative time by augmenting preoperative planning and interfacing with image-guided robotic surgical systems. Study Design Descriptive study of predicted segmentations. Setting Academic institution. Methods We have developed a computational pipeline based on the symmetric normalization registration method that predicts segmentations of anatomic structures in temporal bone CT scans using a labeled atlas. To evaluate accuracy, we created a data set by manually labeling relevant anatomic structures (eg, ossicles, labyrinth, facial nerve, external auditory canal, dura) for 16 deidentified high-resolution cone beam temporal bone CT images. Automated segmentations from this pipeline were compared against ground-truth manual segmentations by using modified Hausdorff distances and Dice scores. Runtimes were documented to determine the computational requirements of this method. Results Modified Hausdorff distances and Dice scores between predicted and ground-truth labels were as follows: malleus (0.100 ± 0.054 mm; Dice, 0.827 ± 0.068), incus (0.100 ± 0.033 mm; Dice, 0.837 ± 0.068), stapes (0.157 ± 0.048 mm; Dice, 0.358 ± 0.100), labyrinth (0.169 ± 0.100 mm; Dice, 0.838 ± 0.060), and facial nerve (0.522 ± 0.278 mm; Dice, 0.567 ± 0.130). A quad-core 16GB RAM workstation completed this segmentation pipeline in 10 minutes. Conclusions We demonstrated submillimeter accuracy for automated segmentation of temporal bone anatomy when compared against hand-segmented ground truth using our template registration pipeline. This method is not dependent on the training data volume that plagues many complex deep learning models. Favorable runtime and low computational requirements underscore this method’s translational potential.

Download Full-text

Clinical Feasibility of Multiplanar Reconstruction Images of Temporal Bone CT in the Diagnosis of Temporal Bone Fracture with Otic-Capsule-Sparing Facial Nerve Paralysis

Indian Journal of Otolaryngology and Head & Neck Surgery ◽

10.1007/s12070-011-0471-8 ◽

2012 ◽

Vol 65 (3) ◽

pp. 219-224 ◽

Cited By ~ 4

Author(s):

J. H. Lim ◽

B. C. Jun ◽

S. W. Song

Keyword(s):

Facial Nerve ◽

Temporal Bone ◽

Bone Fracture ◽

Multiplanar Reconstruction ◽

Facial Nerve Paralysis ◽

Otic Capsule ◽

Temporal Bone Fracture ◽

Nerve Paralysis ◽

Temporal Bone Ct

Download Full-text

Morphometry of the Digastric Ridge and Its Significance for Mastoid Surgeries—A Cadaveric Study

Ear Nose & Throat Journal ◽

10.1177/0145561319870491 ◽

2019 ◽

pp. 014556131987049

Author(s):

David Victor Kumar Irugu ◽

Anup Singh ◽

Rajeev Kumar

Keyword(s):

Facial Nerve ◽

Temporal Bone ◽

Sigmoid Sinus ◽

Risk Of Injury ◽

Temporal Bone Pneumatization ◽

Important Landmark ◽

The Mean ◽

Temporal Bones ◽

Digital Caliper ◽

Reduced Risk

Objective: Digastric ridge (DR) is an important landmark to locate facial nerve (FN) and sigmoid sinus for mastoid surgeries and transmastoid approaches. We aim to look for the effect of temporal bone pneumatization on the morphometry of the DR and its relation to the adjoining structures. Methods: Temporal bones were harvested from unclaimed cadavers after the approval of the ethical committee. The dissection of the temporal bones was performed under a microscope, and the length of the DR and the distance between the mastoid segment of the FN and the anterior end of DR (FN-DR distance) were measured using a digital caliper. Stata version 14.0 was used to perform the statistical calculations. Results: Ninety-three temporal bones were microdissected (right:left = 47:46; well pneumatized:poorly pneumatized = 58:35). Mean length of the DR was 17.1 mm and was significantly longer in well-pneumatized bones ( P = .0000). The mean distance between the anterior end of the digastric ridge and the mastoid part of the facial nerve was 4 mm. The distance was significantly more in well-pneumatized bones. Conclusion: Prominence and the length of the DR, as well as the FN-DR distance, are significantly more in well-pneumatized bones compared to poorly pneumatized bones. This finding has potential surgical implications with reduced risk of injury to the FN resulting from a conspicuous DR in well-pneumatized bones.

Download Full-text

Analysis of translational errors in frame-based and frameless cranial radiosurgery using an anthropomorphic phantom

Radiologia Brasileira ◽

10.1590/0100-3984.2015.0053 ◽

2016 ◽

Vol 49 (2) ◽

pp. 98-103 ◽

Cited By ~ 1

Author(s):

Taynná Vernalha Rocha Almeida ◽

Arno Lotar Cordova Junior ◽

Pedro Argolo Piedade ◽

Cintia Mara da Silva ◽

Priscila Marins ◽

...

Keyword(s):

Anthropomorphic Phantom ◽

Setup Errors ◽

Mean Values ◽

Image Guided ◽

Mean Error ◽

Computed Tomography Images ◽

Error Magnitude ◽

Combined Uncertainty ◽

The Mean ◽

Residual Errors

Abstract Objective: To evaluate three-dimensional translational setup errors and residual errors in image-guided radiosurgery, comparing frameless and frame-based techniques, using an anthropomorphic phantom. Materials and Methods: We initially used specific phantoms for the calibration and quality control of the image-guided system. For the hidden target test, we used an Alderson Radiation Therapy (ART)-210 anthropomorphic head phantom, into which we inserted four 5mm metal balls to simulate target treatment volumes. Computed tomography images were the taken with the head phantom properly positioned for frameless and frame-based radiosurgery. Results: For the frameless technique, the mean error magnitude was 0.22 ± 0.04 mm for setup errors and 0.14 ± 0.02 mm for residual errors, the combined uncertainty being 0.28 mm and 0.16 mm, respectively. For the frame-based technique, the mean error magnitude was 0.73 ± 0.14 mm for setup errors and 0.31 ± 0.04 mm for residual errors, the combined uncertainty being 1.15 mm and 0.63 mm, respectively. Conclusion: The mean values, standard deviations, and combined uncertainties showed no evidence of a significant differences between the two techniques when the head phantom ART-210 was used.

Download Full-text