Literature DB >> 32109685

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

Floris Heutink1, Valentin Koch2, Berit Verbist3, Willem Jan van der Woude2, Emmanuel Mylanus1, Wendy Huinck1, Ioannis Sechopoulos4, Marco Caballo5.   

Abstract

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.
Copyright © 2020 Elsevier B.V. All rights reserved.

Entities:  

Keywords:  Cochlea; Convolutional neural networks; Deep learning; Image segmentation; Ultra-high-resolution CT

Mesh:

Year:  2020        PMID: 32109685     DOI: 10.1016/j.cmpb.2020.105387

Source DB:  PubMed          Journal:  Comput Methods Programs Biomed        ISSN: 0169-2607            Impact factor:   5.428


  10 in total

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

Authors:  Yubo Fan; Dongqing Zhang; Rueben Banalagay; Jianing Wang; Jack H Noble; Benoit M Dawant
Journal:  J Med Imaging (Bellingham)       Date:  2021-11-24

2.  Endolymphatic space is age-dependent.

Authors:  Marianne Dieterich; Tatjana Hergenroeder; Rainer Boegle; Johannes Gerb; Emilie Kierig; Sophia Stöcklein; Valerie Kirsch
Journal:  J Neurol       Date:  2022-10-05       Impact factor: 6.682

Review 3.  A review on the application of deep learning for CT reconstruction, bone segmentation and surgical planning in oral and maxillofacial surgery.

Authors:  Jordi Minnema; Anne Ernst; Maureen van Eijnatten; Ruben Pauwels; Tymour Forouzanfar; Kees Joost Batenburg; Jan Wolff
Journal:  Dentomaxillofac Radiol       Date:  2022-05-23       Impact factor: 3.525

Review 4.  Deep Learning Approaches for Automatic Localization in Medical Images.

Authors:  H Alaskar; A Hussain; B Almaslukh; T Vaiyapuri; Z Sbai; Arun Kumar Dubey
Journal:  Comput Intell Neurosci       Date:  2022-06-29

5.  CT-based radiomics in predicting pathological response in non-small cell lung cancer patients receiving neoadjuvant immunotherapy.

Authors:  Qian Lin; Hai Jun Wu; Qi Shi Song; Yu Kai Tang
Journal:  Front Oncol       Date:  2022-10-04       Impact factor: 5.738

6.  IE-Vnet: Deep Learning-Based Segmentation of the Inner Ear's Total Fluid Space.

Authors:  Seyed-Ahmad Ahmadi; Johann Frei; Gerome Vivar; Marianne Dieterich; Valerie Kirsch
Journal:  Front Neurol       Date:  2022-05-11       Impact factor: 4.086

7.  Effect of Filtered Back-Projection Filters to Low-Contrast Object Imaging in Ultra-High-Resolution (UHR) Cone-Beam Computed Tomography (CBCT).

Authors:  Sunghoon Choi; Chang-Woo Seo; Bo Kyung Cha
Journal:  Sensors (Basel)       Date:  2020-11-10       Impact factor: 3.576

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

Authors:  C A Neves; E D Tran; I M Kessler; N H Blevins
Journal:  Sci Rep       Date:  2021-01-08       Impact factor: 4.379

9.  Computational Audiology: New Approaches to Advance Hearing Health Care in the Digital Age.

Authors:  Jan-Willem A Wasmann; Cris P Lanting; Wendy J Huinck; Emmanuel A M Mylanus; Jeroen W M van der Laak; Paul J Govaerts; De Wet Swanepoel; David R Moore; Dennis L Barbour
Journal:  Ear Hear       Date:  2021 Nov-Dec 01       Impact factor: 3.570

10.  Practicable assessment of cochlear size and shape from clinical CT images.

Authors:  Andrew H Gee; Yufeng Zhao; Graham M Treece; Manohar L Bance
Journal:  Sci Rep       Date:  2021-02-10       Impact factor: 4.379

  10 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.