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

Pathological Voice Detection and Classification Based on Multimodal Transmission Network

OBJECTIVES

Describing pronunciation features from multiple perspectives can help doctors accurately diagnose the pathological type of a patient's voice. According to the two modal information of sound signal and electroglottography (EGG) signal, this paper proposes a pathological voice detection and classification algorithm based on multimodal transmission network.

METHODS

Firstly, we used the short-time Fourier transform (STFT) to map the features of the two signals, and designed the Mel filter to obtain the Mel spectogram. Then, the constructed multimodal transmission network extracted features from Mel spectogram and applied Multimodal Transfer Module (MMTM) module. Finally, the fusion layer can integrate multimodal information, and the full connection layer diagnoses and classifies voice pathology according to the fused features.

RESULTS

The experiment was based on 1179 subjects in Saarbrücken voice database (SVD), and the average accuracy, recall, specificity and F1 score of pathological voice classification reached 98.02%, 98.23%, 97.82% and 97.95% respectively. Compared with other algorithms, the classification accuracy is significantly improved.

CONCLUSIONS

The proposed model can integrate multiple modal information to obtain more comprehensive and stable voice features and improve the accuracy of pathological voice classification. Future research will further explore in reducing the time-consuming and complexity of the model.

Reproducible Machine Learning-Based Voice Pathology Detection: Introducing the Pitch Difference Feature

PURPOSE

We introduce a novel methodology for voice pathology detection using the publicly available Saarbrücken Voice Database and a robust feature set combining commonly used acoustic handcrafted features with two novel ones: pitch difference (relative variation in fundamental frequency) and NaN feature (failed fundamental frequency estimation).

METHODS

We evaluate six machine learning (ML) algorithms-support vector machine, k-nearest neighbors, naive Bayes, decision tree, random forest, and AdaBoost-using grid search for feasible hyperparameters and 20 480 different feature subsets. Top 1000 classification models-feature subset combinations for each ML algorithm are validated with repeated stratified cross-validation. To address class imbalance, we apply k-means synthetic minority oversampling technique to augment the training data.

RESULTS

Our approach achieves 85.61%, 84.69%, and 85.22% unweighted average recall for females, males, and combined results, respectively. We intentionally omit accuracy as it is a highly biased metric for imbalanced data.

CONCLUSION

Our study demonstrates that by following the proposed methodology and feature engineering, there is a potential in detection of various voice pathologies using ML models applied to the simplest vocal task, a sustained utterance of the vowel /a:/. To enable easier use of our methodology and to support our claims, we provide a publicly available GitHub repository with DOI 10.5281/zenodo.13771573. Finally, we provide a REFORMS checklist to enhance readability, reproducibility, and justification of our approach.

Reliability and Validity of GRBASzero in Clinical Environments

PURPOSE

This study aimed to evaluate the reliability and validity of GRBASzero in a real clinical setting.

METHODS

The reliability and validity of GRBASzero were assessed using two independent datasets. Dataset 1 included 283 outpatients who underwent both GRBASzero assessment and human expert evaluation. Dataset 2 from the Perceptual Voice Qualities Database comprised 287 voice samples that underwent evaluation by GRBASzero and were subsequently compared with GRBAS (Grade, Roughness, Breathiness, Asthenicity, Strain) ratings provided by human experts. The reliability of GRBASzero was assessed using Fleiss Kappa, while the validity of GRBASzero was examined using the intraclass correlation coefficient.

RESULTS

In dataset 1, the test-retest reliability of GRBASzero was poor, with the consistency of features A and S approaching random allocation. Consistency analysis with human experts showed a poor agreement for all features except for B. In dataset 2, there was also a poor agreement between GRBASzero and human experts.

CONCLUSION

The reliability and validity of GRBASzero in a real clinical environment are poor and do not meet the requirements for clinical testing, indicating the need for further optimization and improvement.

Toward Real-World Voice Disorder Classification

OBJECTIVE

Voice disorders significantly compromise individuals' ability to speak in their daily lives. Without early diagnosis and treatment, these disorders may deteriorate drastically. Thus, automatic classification systems at home are desirable for people who are inaccessible to clinical disease assessments. However, the performance of such systems may be weakened due to the constrained resources and domain mismatch between the clinical data and noisy real-world data.

METHODS

This study develops a compact and domain-robust voice disorder classification system to identify the utterances of health, neoplasm, and benign structural diseases. Our proposed system utilizes a feature extractor model composed of factorized convolutional neural networks and subsequently deploys domain adversarial training to reconcile the domain mismatch by extracting domain-invariant features.

RESULTS

The results show that the unweighted average recall in the noisy real-world domain improved by 13% and remained at 80% in the clinic domain with only slight degradation. The domain mismatch was effectively eliminated. Moreover, the proposed system reduced the usage of both memory and computation by over 73.9%.

CONCLUSION

By deploying factorized convolutional neural networks and domain adversarial training, domain-invariant features can be derived for voice disorder classification with limited resources. The promising results confirm that the proposed system can significantly reduce resource consumption and improve classification accuracy by considering the domain mismatch.

SIGNIFICANCE

To the best of our knowledge, this is the first study that jointly considers real-world model compression and noise-robustness issues in voice disorder classification. The proposed system is intended for application to embedded systems with limited resources.

A review on voice pathology: Taxonomy, diagnosis, medical procedures and detection techniques, open challenges, limitations, and recommendations for future directions

Speech is a primary means of human communication and one of the most basic features of human conduct. Voice is an important part of its subsystems. A speech disorder is a condition that affects the ability of a person to speak normally, which occasionally results in voice impairment with psychological and emotional consequences. Early detection of voice problems is a crucial factor. Computer-based procedures are less costly and easier to administer for such purposes than traditional methods. This study highlights the following issues: recent studies, methods of voice pathology detection, machine learning and deep learning (DL) methods used in data classification, main datasets utilized, and the role of Internet of things (IoT) systems employed in voice pathology diagnosis. Moreover, this study presents different applications, open challenges, and recommendations for future directions of IoT systems and artificial intelligence (AI) approaches in the voice pathology diagnosis. Finally, this study highlights some limitations of voice pathology datasets in comparison with the role of IoT in the healthcare sector, which shows the urgent need to provide efficient approaches and easy and ideal medical diagnostic procedures and treatments of disease identification for doctors and patients. This review covered voice pathology taxonomy, detection techniques, open challenges, limitations, and recommendations for future directions to provide a clear background for doctors and patients. Standard databases, including the Massachusetts Eye and Ear Infirmary, Saarbruecken Voice Database, and the Arabic Voice Pathology Database, were used in most articles reviewed in this article. The classes, features, and main purpose for voice pathology identification are also highlighted. This study focuses on the extraction of voice pathology features, especially speech analysis, extends feature vectors comprising static and dynamic features, and converts these extended feature vectors into solid vectors before passing them to the recognizer.

Lightweight Deep Learning Model for Assessment of Substitution Voicing and Speech after Laryngeal Carcinoma Surgery

Laryngeal carcinoma is the most common malignant tumor of the upper respiratory tract. Total laryngectomy provides complete and permanent detachment of the upper and lower airways that causes the loss of voice, leading to a patient's inability to verbally communicate in the postoperative period. This paper aims to exploit modern areas of deep learning research to objectively classify, extract and measure the substitution voicing after laryngeal oncosurgery from the audio signal. We propose using well-known convolutional neural networks (CNNs) applied for image classification for the analysis of voice audio signal. Our approach takes an input of Mel-frequency spectrogram (MFCC) as an input of deep neural network architecture. A database of digital speech recordings of 367 male subjects (279 normal speech samples and 88 pathological speech samples) was used. Our approach has shown the best true-positive rate of any of the compared state-of-the-art approaches, achieving an overall accuracy of 89.47%.

Voice pathology detection and classification from speech signals and EGG signals based on a multimodal fusion method

Automatic voice pathology detection and classification plays an important role in the diagnosis and prevention of voice disorders. To accurately describe the pronunciation characteristics of patients with dysarthria and improve the effect of pathological voice detection, this study proposes a pathological voice detection method based on a multi-modal network structure. First, speech signals and electroglottography (EGG) signals are mapped from the time domain to the frequency domain spectrogram via a short-time Fourier transform (STFT). The Mel filter bank acts on the spectrogram to enhance the signal's harmonics and denoise. Second, a pre-trained convolutional neural network (CNN) is used as the backbone network to extract sound state features and vocal cord vibration features from the two signals. To obtain a better classification effect, the fused features are input into the long short-term memory (LSTM) network for voice feature selection and enhancement. The proposed system achieves 95.73% for accuracy with 96.10% F1-score and 96.73% recall using the Saarbrucken Voice Database (SVD); thus, enabling a new method for pathological speech detection.