Diagnosis of hearing deficiency using EEG based AEP signals: CWT and improved-VGG16 pipeline.

Md Nahidul Islam,Mamunur Rashid,Fahmid Al Farid,Norizam Sulaiman,Salem A Alyami,Mohammad Ali Moni,Anwar P P Abdul Majeed,Jia Uddin

doi:10.7717/peerj-cs.638

Abstract

Hearing deficiency is the world’s most common sensation of impairment and impedes human communication and learning. Early and precise hearing diagnosis using electroencephalogram (EEG) is referred to as the optimum strategy to deal with this issue. Among a wide range of EEG control signals, the most relevant modality for hearing loss diagnosis is auditory evoked potential (AEP) which is produced in the brain’s cortex area through an auditory stimulus. This study aims to develop a robust intelligent auditory sensation system utilizing a pre-train deep learning framework by analyzing and evaluating the functional reliability of the hearing based on the AEP response. First, the raw AEP data is transformed into time-frequency images through the wavelet transformation. Then, lower-level functionality is eliminated using a pre-trained network. Here, an improved-VGG16 architecture has been designed based on removing some convolutional layers and adding new layers in the fully connected block. Subsequently, the higher levels of the neural network architecture are fine-tuned using the labelled time-frequency images. Finally, the proposed method’s performance has been validated by a reputed publicly available AEP dataset, recorded from sixteen subjects when they have heard specific auditory stimuli in the left or right ear. The proposed method outperforms the state-of-art studies by improving the classification accuracy to 96.87% (from 57.375%), which indicates that the proposed improved-VGG16 architecture can significantly deal with AEP response in early hearing loss diagnosis.

Highlights

Hearing deficiency is the widespread form of human sensory disability; it is the partial or complete inability to listen to the ear’s sound
We have presented a hearing deficiency identification system based on deep convolutional neural networks (CNNs), where a transfer learning strategy has been used to improve the training process
The main goal of this study is to enhance the performance for detecting the hearing condition with a concise decision window, so that we can efficiently use this system in reallife application

Summary

Introduction

Hearing deficiency is the widespread form of human sensory disability; it is the partial or complete inability to listen to the ear’s sound. An early and effective hearing screening test is essential for address the vast population concern. That helps to reduce the hearing deficiency by taking necessary steps at an appropriate time. Conventional listening tests and audiograms appear to be subjective assessments that significantly demand medical and health services. The audiogram reflects the hearing threshold across the speech frequency spectrum, usually between 125 and 8,000 Hz. The traditional hearing impairment testing technique is very time-consuming, takes sufficient clinical time and expertise to interpret and maintain since it requires the person to respond directly. In the application of hearing aid, other issues, such as hearing loss’s consequence (Holmes, Kitterick & Summerfield, 2017), the circumstances of the auditory stimulus (such as the background noise of the stimulus, locations of the stimulus (Das, Bertrand & Francart, 2018; Das et al, 2016) ), attentionaltering techniques is still an open question

Objectives

Findings

Discussion

Conclusion