Sensorineural hearing loss classification via deep-HLNet and few-shot learning
- PDF / 1,479,412 Bytes
- 14 Pages / 439.37 x 666.142 pts Page_size
- 74 Downloads / 145 Views
Sensorineural hearing loss classification via deep-HLNet and few-shot learning Xi Chen 1,2 & Qinghua Zhou 2 & Rushi Lan 1 & Shui-Hua Wang 2 & Yu-Dong Zhang 2 Xiaonan Luo 1
&
Received: 11 May 2020 / Revised: 11 August 2020 / Accepted: 21 August 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
We propose a new method for hearing loss classification from magnetic resonance image (MRI), which can automatically detect tissue-specific features in a given MRI. Sensorineural hearing loss (SHNL) is highly prevalent in our society. Early diagnosis and intervention have a profound impact on patient outcomes. A solution to provide early diagnosis is the use of automated diagnostic systems. In this study, we propose a novel Deep-HLNet framework, based on few-shot learning, for the automated classification of SNHL. This research involves magnetic resonance (MRI) images from 60 participants of three balanced categories: left-sided SNHL, right-sided SNHL, and healthy controls. A convolutional neural network was employed for feature extraction from individual categories, while a neural network and a comparison classifier strategy constituted a triclassifier for SNHL classification. In terms of experiment results and practicability of the algorithm, the classification performance was significantly better than the standard deep learning methods or other conventional methods, with an overall accuracy of 96.62%. Keywords Hearing loss . Few-shot learning . Deep-HLNet
1 Introduction Sensorineural hearing loss (SNHL) is the condition of problematic neural signal transfer from the cochlea towards the auditory cortex. In general, this disease can be congenital or acquired. * Rushi Lan [email protected] * Yu-Dong Zhang [email protected]
1
Guangxi Key Laboratory of Image and Graphic Intelligent Processing, Guilin University of Electronic Technology, Guilin 541004, China
2
Department of Informatics, University of Leicester, Leicester LE1 7RH, UK
Multimedia Tools and Applications
Congenital SNHL is based on heredity and present as malformation of the inner ear at birth. Aging, head trauma, tumors, and exposure to loud noise leads to acquired SNHL. More than half of these causes are preventable but not reversible [24]. Therefore, early detection of hearing loss and appropriate treatment is essential. Currently, machine learning has been recognized as an efficient tool in fields of medicine, like medical images processing [1]. For instance, computer-assisted diagnosis and medical analysis can enhance the abilities of doctors in tasks of medical image detection, recognition, and segmentation. Deep learning algorithms such as convolutional neural networks (CNN) [45], autoencoders (AE), recurrent neural networks (RNN) [12] and generative adversarial networks (GAN) [5] have shown higher performance in many cases compared to traditional algorithms such as support vector machine (SVM) [9, 13, 18, 46, 47], random forest (RF) and decision tree (DT), etc. Deep learning allows automatic and fast feature extr
Data Loading...
