Towards an automatic evaluation of the dysarthria level of patients with Parkinson's disease

Abstract

The m-FDA scale was introduced to assess the dysarthria level of patients with PD. Articulation features extracted from continuous speech signals to create i-vectors were the most accurate to quantify the dysarthria level, with correlations of up to 0.69 between the predicted m-FDA scores and those assigned by the phoniatricians. When the dysarthria levels were estimated considering dedicated speech exercises such as rapid repetition of syllables (DDKs) and read texts, the correlations were 0.64 and 0.57, respectively. In addition, the combination of several feature sets and speech tasks improved the results, which validates the hypothesis about the contribution of information from different tasks and feature sets when assessing dysarthric speech signals. The speaker models seem to be promising to perform individual modeling for monitoring the dysarthria level of PD patients. The proposed approach may help clinicians to make more accurate and timely decisions about the evaluation and therapy associated to the dysarthria level of patients. The proposed approach is a great step towards unobtrusive/ecological evaluations of patients with dysarthric speech without the need of attending medical appointments.

Articulation and Empirical Mode Decomposition Features in Diadochokinetic Exercises for the Speech Assessment of Parkinson’s Disease Patients

“…Similarly to the classification experiments, the highest correlation is obtained with the combination of transition and regularity features (ρ=0.6782). The "strong" correlation obtained is statistically significant, and it is comparable to the obtained in related studies, where the same problem was addressed [14,18].…”

“…The evaluation of PD patients according to the MDS-UPDRS-III scale is suitable to assess general motor impairments of PD patients; however, the deterioration of the communication skills is not properly evaluated because such a scale only considers speech impairments in one of its items. We recently introduced the m-FDA scale, which is focused on speech impairments showed by PD patients and can be administered based on speech recordings [14]. The scale includes several aspects of speech: respiration, lips movement, palate/velum movement, larynx, tongue, monotonicity, and intelligibility.…”

“…Finally, the third feature set considers descriptors extracted from the empirical mode decomposition (EMD), which have been recently used to model dysphonia in PD patients [12,13]. The combination of the articulation features to model the start/stop movement of the vocal folds with the features to model the regularity of the DDK exercises showed to be highly accurate to classify PD patients and HC subjects, and to predict the dysarthria level of the participants, according to a modified version of the Frenchay dysarthria assessment scale (m-FDA), which was proposed recently in [14].…”

Detecting Parkinson’s disease with sustained phonation and speech signals using machine learning techniques

“…Recently, other approaches have been investigated for the detection of PD. [16] studied speech recordings using feature extracted from several dimensions of speech, including phonation, articulation and other human characteristics. To improve the diagnosis of Parkinson's disease, [17] introduced an improved and optimized version of the Crow Search Algorithm.…”

Convolutional Neural Networks and a Transfer Learning Strategy to Classify Parkinson’s Disease from Speech in Three Different Languages

“…In addition, the base models will be trained considering two of the languages instead of only one of them. The trained models will also be evaluated to classify the speech of PD patients in several stages of the disease based on the MDS-UPDRS-III score, or based on their dysarthria severity [21]. Further experiments will also include transfer learning among diseases, for instance training a base model with utterances to classify PD, and use such a model to initialize another one to classify other neurological diseases such as Hungtinton's disease.…”