Segmentation of Cochlear Nerve Based on Particle Swarm Optimization Method

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 490)

Abstract

Sensorineural hearing loss is a hearing impairment happens when there is damage to the inner ear or to the nerve pathways from the internal ear to the brain. Cochlear implants have been developed to help the patients with congenital or acquired hearing loss. The size of the cochlear nerve is a prerequisite for the successful outcome of cochlear implant surgery. Hence, an accurate segmentation of cochlear nerve is a critical assignment in computer-aided diagnosis and surgery planning of cochlear implants. This paper aims at developing a cochlear nerve segmentation approach based on modified particle swarm optimization (PSO). In the proposed approach, a constant adaptive inertia weight based on the kernel density estimation of the image histogram is estimated for fine-tuning the current search space to segment the cochlear nerve. The segmentation results are analyzed both qualitatively and quantitatively based on the performance measures, namely Jaccard index, Dice coefficient, sensitivity, specificity, and accuracy as well. These results indicate that the proposed algorithm performs better compared to standard PSO algorithm in preserving edge details and boundary shape. The proposed method is tested on different slices of eight patients undergone for magnetic resonance imaging in the assessment of giddiness/vertigo or fitness for the cochlear implant. The significance of this work is to segment the cochlear nerve from magnetic resonance (MR) images to assist the radiologists in their diagnosis and for successful cochlear implantation with the scope of developing speech and language, especially in children.

Keywords

Cochlear nerve Cochlear implant Improved PSO segmentation Magnetic resonance (MR) image 

Notes

Acknowledgements

We express our gratitude to Dr. R. Rajeshwaran for his helpful data on demonstrative detail of internal ear MR images. Likewise, authors might want to thank the Sri Ramachandra Medical Center, Porur, Chennai, India, for giving the MR images.

References

  1. 1.
    Adunka OF, Roush PA, Teagle HF, Brown CJ, Zdanski CJ, Jewells V, Buchman CA (2006) Internal auditory canal morphology in children with cochlear nerve deficiency. Otol Neurotology 27(6):793–801CrossRefGoogle Scholar
  2. 2.
    Kim BG, Chung HJ, Park JJ, Park S, Kim SH, Choi JY (2013) Correlation of cochlear nerve size and auditory performance after cochlear implantation in postlingually deaf patients. JAMA Otolaryngol Head Neck Surg 139(6):604–609CrossRefGoogle Scholar
  3. 3.
    Lemieux G, Krakow K, Woermann F (1999) Fast, accurate, and reproducible automatic segmentation of the brain in weighted volume MRI data. Magn Reson Med 42:127–135CrossRefGoogle Scholar
  4. 4.
    Tang H, Wu E, Ma Q, Gallagher D, Perera G, Zhuang T (2000) MRI brain image segmentation by multi-resolution edge detection and region selection. Comput Med Imag Graph 24:349–357CrossRefGoogle Scholar
  5. 5.
    Liew AWC, Yan H (2003) An adaptive spatial fuzzy clustering algorithm for 3-D MR image segmentation. IEEE Trans Med Imag 22:1063–1075CrossRefGoogle Scholar
  6. 6.
    Kennedy J, Eberhart R (1995) Particle swarm optimization. In: Proceedings of the IEEE international conference on neural networks, Perth, Australia, vol 4, no 2, pp 1942–1948Google Scholar
  7. 7.
    Lee C-Y, Leou J-J, Hsiao H-H (2012) Saliency-directed color image segmentation using modified particle swarm optimization. Signal Process 92:1–18CrossRefGoogle Scholar
  8. 8.
    Wood JC, Johnson KM (1999) Wavelet packet denoising of magnetic resonance images: importance of Rician noise at low SNR. Magn Reson Med 41(3):631–635CrossRefGoogle Scholar
  9. 9.
    Jeevakala S, Brintha Therese A (2016) Non local means filter based Rician noise removal of MR images. Int J Pure Appl Math 109(5):133–139Google Scholar
  10. 10.
    Meijering EHW, Niessen WJ, Viergever MA (2001) Quantitative evaluation of convolution-based methods for medical image interpolation. Med Image Anal 5:111–126CrossRefGoogle Scholar
  11. 11.
    Li H, He H, Wen Y (2015) Dynamic particle swarm optimization and K-means clustering algorithm for image segmentation. Opt-Int J Light Electron Opt 126(24):4817–4848CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  1. 1.School of Electronics EngineeringVIT UniversityChennaiIndia

Personalised recommendations