Towards the identification of Idiopathic Parkinson’s Disease from the speech. New articulatory kinetic biomarkers

Abstract

Although a large amount of acoustic indicators have already been proposed in the literature to evaluate the hypokinetic dysarthria of people with Parkinson’s Disease, the goal of this work is to identify and interpret new reliable and complementary articulatory biomarkers that could be applied to predict/evaluate Parkinson’s Disease from a diadochokinetic test, contributing to the possibility of a further multidimensional analysis of the speech of parkinsonian patients. The new biomarkers proposed are based on the kinetic behaviour of the envelope trace, which is directly linked with the articulatory dysfunctions introduced by the disease since the early stages. The interest of these new articulatory indicators stands on their easiness of identification and interpretation, and their potential to be translated into computer based automatic methods to screen the disease from the speech. Throughout this paper, the accuracy provided by these acoustic kinetic biomarkers is compared with the one obtained with a baseline system based on speaker identification techniques. Results show accuracies around 85% that are in line with those obtained with the complex state of the art speaker recognition techniques, but with an easier physical interpretation, which open the possibility to be transferred to a clinical setting.

Study of the Automatic Detection of Parkison’s Disease Based on Speaker Recognition Technologies and Allophonic Distillation

“…However, the sequence of syllables employed in the recordings (/pa-ta-ka/) contains only plosives. This fact can be explained by the inclination of many patients to replace plosives with fricatives [4], phenomenon (known as spirantization) which can be better detected with a classifier more influenced by the acoustic characteristics of fricatives or plosives.…”

“…More recent works explore the automatic detection or assessment of PD by analyzing different types of segments of speech, the velocity or acceleration of articulators, specific transitions or the evolution of formants [2]- [4]. As an example, work [2] reports 80% of accuracy in PD detection using acoustic features extracted from only vowels in monologues.…”

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

“…Some preliminary findings suggest that this might be the case. For example, it has been found that voiced segments tend to be longer, while stop silences produced by the closures before bursts tend to disappear, in the speech of PD patients compared to control speakers 7,20 . These observations are in direct relationship with misarticulation phenomena common in some dysarthrias by which plosives are produced as fricatives, where the frication noise is not necessarily preceded by a closure.…”

“…On the other hand, differences in the slopes and variability of the formant frequencies between patients and controls (and especially between the vowel space areas (VSA) of both groups) have been reported 21–27 . Based on that, recent studies propose automatic systems to detect or assess PD making use of the articulatory aspects of speech and advanced signal processing techniques, suggesting that speech processing can derive powerful indicators of imprecise consonant articulation in PD-related dysarthria 20,21,28–30 . The accuracy in PD detection of these works, as in most of the works in the literature, does not exceed 90%, although it is difficult to compare the performance of the different methodologies since each study uses a different corpus and evaluates its results following a different procedure.…”

Feature Representation of Pathophysiology of Parkinsonian Dysarthria

“…This indicates that changes in syllables is more difficult than repeating the same syllable. However, it was unexpected to see /ta/ on the list instead of /ka/ for Spanish speakers [25].…”