Class-Imbalanced Voice Pathology Detection and Classification Using Fuzzy Cluster Oversampling Method

Ziqi Fan,Changwei Zhou,Xiaojun Zhang,Yuanbo Wu,Zhi Tao

doi:10.3390/app11083450

Ziqi Fan, Changwei Zhou + Show 3 more

Open Access

https://doi.org/10.3390/app11083450

Copy DOI

Journal: Applied Sciences	Publication Date: Apr 12, 2021
Citations: 18	License type: CC BY 4.0

Affiliation: Soochow University

Abstract

The Massachusetts Eye and Ear Infirmary (MEEI) database is an international-standard training database for voice pathology detection (VPD) systems. However, there is a class-imbalanced distribution in normal and pathological voice samples and different types of pathological voice samples in the MEEI database. This study aimed to develop a VPD system that uses the fuzzy clustering synthetic minority oversampling technique algorithm (FC-SMOTE) to automatically detect and classify four types of pathological voices in a multi-class imbalanced database. The proposed FC-SMOTE algorithm processes the initial class-imbalanced dataset. A set of machine learning models was evaluated and validated using the resulting class-balanced dataset as an input. The effectiveness of the VPD system with FC-SMOTE was further verified by an external validation set and another pathological voice database (Saarbruecken Voice Database (SVD)). The experimental results show that, in the multi-classification of pathological voice for the class-imbalanced dataset, the method we propose can significantly improve the diagnostic accuracy. Meanwhile, FC-SMOTE outperforms the traditional imbalanced data oversampling algorithms, and it is preferred for imbalanced voice diagnosis in practical applications.

Highlights

Traditional pathological voice detection mainly depends on experienced clinicians or laryngoscopes to observe the vocal cord structure [1], which is subjective and invasive
This paper proposes a voice pathology detection (VPD) system combined with an FC-synthetic minority oversampling technique (SMOTE) imbalanced learning algorithm
The effectiveness of FC-SMOTE is tested on an external validation set of the Massachusetts Eye and Ear Infirmary (MEEI) database

Summary

Introduction

Traditional pathological voice detection mainly depends on experienced clinicians or laryngoscopes to observe the vocal cord structure [1], which is subjective and invasive. Different features are extracted from signals to build VPD systems that automatically detect pathological voices Most of these studies have experimented with the Massachusetts Eye and Ear Infirmary (MEEI) database [5], which has become one of the standard databases for VPD systems [6]. Taking the Acc as the evaluation result of the classifier makes the pathological voice detection model’s performance better than it is. This is because learning with imbalanced datasets usually results in a biased classifier that obtains a higher detection accuracy in majority classes and a lower one in minority classes [7].

Methods

Results

Discussion

Conclusion