Phonetic relevance and phonemic grouping of speech in the automatic detection of Parkinson’s Disease

Moro-Velázquez, Laureano; García, Jorge Andrés Gómez; Godino-Llorente, Juan Ignacio; Grandas-Pérez, F.; Shattuck‐Hufnagel, Stefanie; Yagüe-Jiménez, Virginia; Dehak, Najim

doi:10.1038/s41598-019-55271-y

Cited by 42 publications

(36 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The accuracy of the different cepstral feature sets constructed was evaluated in pathological voice detection using voice signals representing the vowel /a/ from six voice databases: the Hospital Universitario Príncipe de Asturias (HUPA) [51] database, the Neurovoz [53] database, the PC-GITA [54] database, the hyperkinetic dysphonia (HkD) and laryngitis subsets of the Saarbrücken Voice Disorders (SVD) [55], [56] database, and the heart failure (HF) database [57]. Most previous studies in voice pathology detection are based on voice recordings of sustained vowels, particularly representing the vowel /a/ [5], [58], [59].…”

Section: A Considered Voice Pathology Databasesmentioning

confidence: 99%

“…2) The Neurovoz database: This corpus consists of voice signals representing the vowel /a/ produced by 110 Parkinsonian patients and by 93 healthy speakers whose mother tongue is Castillian Spanish [53]. The signals were recorded with a sampling frequency of 44.1 kHz.…”

Section: A Considered Voice Pathology Databasesmentioning

confidence: 99%

“…The signals were recorded with a sampling frequency of 44.1 kHz. More details about the Neurovoz database can be found in [53].…”

Section: A Considered Voice Pathology Databasesmentioning

confidence: 99%

See 2 more Smart Citations

A Comparison of Cepstral Features in the Detection of Pathological Voices by Varying the Input and Filterbank of the Cepstrum Computation

Reddy¹,

Alku

2021

IEEE Access

View full text Add to dashboard Cite

Automatic voice pathology detection enables objective assessment of pathologies that affect the voice production mechanism. Detection systems have been developed using the traditional pipeline approach (consisting of the feature extraction part and the detection part) and using the modern deep learning -based end-to-end approach. Due to the lack of vast amounts of training data in the study area of pathological voice, the former approach is still a valid choice. In the existing detection systems based on the traditional pipeline approach, the mel-frequency cepstral coefficient (MFCC) features can be regarded as the defacto standard feature set. In this study, automatic voice pathology detection is investigated by comparing the performance of various MFCC variants derived by considering two factors: the input and the filterbank in the cepstrum computation. For the first factor, three inputs (the voice signal, the glottal source and the vocal tract) are compared. The glottal source and the vocal tract are estimated using the quasi-closed phase glottal inverse filtering method. For the second factor, the mel-frequency and linear-frequency filterbanks are compared. Experiments were conducted separately using six databases consisting of voices produced by speakers suffering from one of four disorders (dysphonia, Parkinson's disease, laryngitis, or heart failure) and by healthy speakers. Support vector machine (SVM) was used as the classifier. The results show that by combining mel-and linear-frequency cepstral coefficients derived from the glottal source and vocal tract, better overall detection accuracy was obtained compared to the defacto MFCC features derived from the voice signal. Furthermore, this combination provided comparable or better performance than four existing cepstral feature extraction techniques in clean and high signal-to-noise ratio (SNR) conditions.INDEX TERMS Voice disorders, glottal inverse filtering, support vector machine, cepstral coefficients.

show abstract

Section: A Considered Voice Pathology Databasesmentioning

confidence: 99%

Section: A Considered Voice Pathology Databasesmentioning

confidence: 99%

See 1 more Smart Citation

A Comparison of Cepstral Features in the Detection of Pathological Voices by Varying the Input and Filterbank of the Cepstrum Computation

Reddy¹,

Alku

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Its expansion rate is estimated to be 1.5% approximately for people aged over 65 years [3]. The Clinic pathological studies show that up to 25% of the patients with PD are diagnosed incorrectly [4], The accuracy of clinical diagnosis can reach approximately 90% within a period of 2 years and 9 months [5]. Diagnosing PD is rather difficult, up till now there is no blood test that can reveal whether a person has a PD or not.…”

Section: Introductionmentioning

confidence: 99%

An optimized RNN-LSTM approach for parkinson’s disease early detection using speech features

2021

View full text Add to dashboard Cite

Parkinson's disease (PD) is the second most common neurodegenerative disorder disease right after Alzheimer's and the most common movement disorder for elderly people. It is characterized as a progressive loss of muscle control, which leads to trembling characterized by uncontrollable shaking, or (tremors) in different parts of the body. In recent years, deep learning (DL) models achieved significant progress in automatic speech recognition, however, limited studies addressed the problem of distinguishing people with PD for further clinical diagnosis. In this paper, an approach for the early detection of patients with PD using speech features was proposed, a recurrent neural network (RNN) with long short-term memory (LSTM) is applied with the batch normalization layer and adaptive moment estimation (ADAM) optimization algorithm used after the network hidden layers to improve the classification performance. The proposed approach is applied with 2 benchmark datasets of speech features for patients with PD and healthy control subjects. The proposed approach achieved an accuracy of 95.8% and MCC=92.04% for the testing dataset. In future work, we aim to increase the voice features that will be worked on and consider using handwriting kinematic features.

show abstract

“…The physiological motivation behind the TRs analysis is the effect of the lack of coordination, typical of PDPs, in the use of the source glottal [16], [17]. Indeed, the direct visualization of the vocal fold vibration by video laryngoscopy [11] revealed incomplete glottis closure due to impaired vocal fold abduction and bowing.…”

Section: Introductionmentioning

confidence: 99%

Speech Impairment in Parkinson’s Disease: Acoustic Analysis of Unvoiced Consonants in Italian Native Speakers

Amato¹,

Borzì

Olmo

et al. 2021

IEEE Access

View full text Add to dashboard Cite

The study of the influence of Parkinson's Disease (PD) on vocal signals has received much attention over the last decades. Increasing interest has been devoted to articulation and acoustic characterization of different phonemes.Method: In this study we propose the analysis of the Transition Regions (TR) of specific phonetic groups to model the loss of motor control and the difficulty to start/stop movements, typical of PD patients. For this purpose, we extracted 60 features from pre-processed vocal signals and used them as input to several machine learning models. We employed two data sets, containing samples from Italian native speakers, for training and testing. The first dataset -28 PD patients and 22 Healthy Control (HC) -included recordings in optimal conditions, while in the second one -26 PD patients and 18 HC-signals were collected at home, using non-professional microphones. Results: We optimized two support vector machine models for the application in controlled noise conditions and home environments, achieving 98% ± 1.1 and 88% ± 2.8 accuracy in 10-fold cross-validation, respectively. Conclusion: This study confirms the high capability of the TRs to discriminate between PD patients and healthy controls, and the feasibility of automatic PD assessment using voice recordings. Moreover, the promising performance of the implemented model discloses the option of voice processing using low-cost devices and domestic recordings, possibly self-managed by the patients themselves.

show abstract

Phonetic relevance and phonemic grouping of speech in the automatic detection of Parkinson’s Disease

Cited by 42 publications

References 35 publications

A Comparison of Cepstral Features in the Detection of Pathological Voices by Varying the Input and Filterbank of the Cepstrum Computation

A Comparison of Cepstral Features in the Detection of Pathological Voices by Varying the Input and Filterbank of the Cepstrum Computation

An optimized RNN-LSTM approach for parkinson’s disease early detection using speech features

Speech Impairment in Parkinson’s Disease: Acoustic Analysis of Unvoiced Consonants in Italian Native Speakers

Contact Info

Product

Resources

About