Vocal signal analysis in patients affected by Multiple Sclerosis

Vizza, Patrizia; Mirarchi, Domenico; Tradigo, Giuseppe; Redavide, Maria; Bossio, Roberto Bruno; Veltri, Pierangelo

doi:10.1016/j.procs.2017.05.092

Cited by 20 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of those are used for assessment of voice quality [ 20 ]. Specifically, we took into account classical features such as duration of the file in seconds, fundamental frequency F0 and its standard deviation, Subharmonics and Harmonics information, Jitter [ 21 ], Shimmer [ 22 ], Formants information, Intensity, Speech Rate, Signal to Noise Ratio, Long-Term average spectrum (LTAS) [ 23 ], Root Mean Square energy (RMS), and some spectral shape information. In particular, the list of the 24 features extracted is: Duration in seconds : it is the length of the voice file in seconds; Mean F0 : it is the mean of the Fundamental frequency F0; STD F0 : it is the standard deviation of the Fundamental frequency F0; Subharmonics to Harmonic ratio (SHR) : amplitude ratio between subharmonics and harmonics according to [ 24 ]; Subharmonics pitch : the fundamental frequency estimate introduced in [ 24 ] for impaired speakers; Local jitter : parameter of frequency variation from cycle to cycle; Absolute jitter (accessed on 10 May 2021) : it is the average absolute difference between consecutive periods in seconds; RAP jitter (accessed on 10 May 2021) : Relative Average Perturbation, the average absolute difference between a period and the average of it and its two neighbours, divided by the average period; Local shimmer : amplitude variation of the sound wave [ 25 ]; F1 mean : it is the first formant (accessed on 10 May 2021); F2 mean : it is the second formant; F3 mean : it is the third formant; F4 mean : it is the fourth formant; Formant dispersion : it is the difference between F4 and F1 divided by 3; Mean intensity (accessed on 10 May 2021) : the mean (in dB) of the intensity values of the frames within a specified time domain; Speech rate : number of syllables divided by file duration in seconds; Signal to Noise ratio (SNR) ; LTAS (accessed on 10 May 2021) : it is the mean of logarithmic power spectral density as a function of frequency, computed over the entire domain frequency (from 0 Hz to 5000 Hz); LTAS slope ; LTAS standard deviation ; RMS energy : root-mean-square of energy; Spectrum centre of gravity (SCG) (accessed on 10 May 2021) : it is the average of frequency over the entire spectrum, weighted by the power spectrum; Spectrum standard deviation (accessed on 10 May 2021) : it is the variance of the frequencies in the spectrum; Band Energy difference (accessed on 10 May 2021) : it is the ratio between the average power over low (between 0 Hz and 500 Hz) and high (between 500 Hz and 4000 Hz) frequency bins in decibel scale; …”

Section: Methodsmentioning

confidence: 99%

Section: Second Experimentsmentioning

confidence: 99%

See 1 more Smart Citation

Optimising Speaker-Dependent Feature Extraction Parameters to Improve Automatic Speech Recognition Performance for People with Dysarthria

Marini

Vanello

Fanucci

2021

Sensors

View full text Add to dashboard Cite

Within the field of Automatic Speech Recognition (ASR) systems, facing impaired speech is a big challenge because standard approaches are ineffective in the presence of dysarthria. The first aim of our work is to confirm the effectiveness of a new speech analysis technique for speakers with dysarthria. This new approach exploits the fine-tuning of the size and shift parameters of the spectral analysis window used to compute the initial short-time Fourier transform, to improve the performance of a speaker-dependent ASR system. The second aim is to define if there exists a correlation among the speaker’s voice features and the optimal window and shift parameters that minimises the error of an ASR system, for that specific speaker. For our experiments, we used both impaired and unimpaired Italian speech. Specifically, we used 30 speakers with dysarthria from the IDEA database and 10 professional speakers from the CLIPS database. Both databases are freely available. The results confirm that, if a standard ASR system performs poorly with a speaker with dysarthria, it can be improved by using the new speech analysis. Otherwise, the new approach is ineffective in cases of unimpaired and low impaired speech. Furthermore, there exists a correlation between some speaker’s voice features and their optimal parameters.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Second Experimentsmentioning

confidence: 99%

Optimising Speaker-Dependent Feature Extraction Parameters to Improve Automatic Speech Recognition Performance for People with Dysarthria

Marini

Vanello

Fanucci

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Speech abnormalities can be studied by analyzing several parameters obtained from vocal signal processing with the aim to describe the voice in a quantitative way and to identify patterns [37]. Two types of methods have been used to extract the parameters needed to furnish a more detailed analysis for the evaluation and identification of possible correlations between vocal signals and MS disease.…”

Section: Methodsmentioning

confidence: 99%

On the Use of Voice Signals for Studying Sclerosis Disease

et al. 2017

Self Cite

View full text Add to dashboard Cite

Abstract:Multiple sclerosis (MS) is a chronic demyelinating autoimmune disease affecting the central nervous system. One of its manifestations concerns impaired speech, also known as dysarthria. In many cases, a proper speech evaluation can play an important role in the diagnosis of MS. The identification of abnormal voice patterns can provide valid support for a physician in the diagnosing and monitoring of this neurological disease. In this paper, we present a method for vocal signal analysis in patients affected by MS. The goal is to identify the dysarthria in MS patients to perform an early diagnosis of the disease and to monitor its progress. The proposed method provides the acquisition and analysis of vocal signals, aiming to perform feature extraction and to identify relevant patterns useful to impaired speech associated with MS. This method integrates two well-known methodologies, acoustic analysis and vowel metric methodology, to better define pathological compared to healthy voices. As a result, this method provides patterns that could be useful indicators for physicians in identifying patients affected by MS. Moreover, the proposed procedure could be a valid support in early diagnosis as well as in monitoring treatment success, thus improving a patient's life quality.

show abstract

“…Studies showed a statistically significant correlation between dysphonic symptoms and MS, and the odds for having MS were 2.2 times higher if dysphonic symptoms were present with high jitter and shimmer values as well as high soft phonation index (an indicator of vocal fold adduction; high values correlate with incomplete adduction of the vocal fold [41]) values [42,43]. The objective acoustic analysis of speech seems to be more sensitive for discrimination between affected patients and healthy controls (90% accuracy) than experienced raters (35% accuracy) are, and thus could be used as a biomarker for diagnosis and the monitoring of disease progression [13,35,44,45].…”

Section: Brainstemmentioning

confidence: 99%

Digital Biomarkers in Multiple Sclerosis

et al. 2021

View full text Add to dashboard Cite

For incurable diseases, such as multiple sclerosis (MS), the prevention of progression and the preservation of quality of life play a crucial role over the entire therapy period. In MS, patients tend to become ill at a younger age and are so variable in terms of their disease course that there is no standard therapy. Therefore, it is necessary to enable a therapy that is as personalized as possible and to respond promptly to any changes, whether with noticeable symptoms or symptomless. Here, measurable parameters of biological processes can be used, which provide good information with regard to prognostic and diagnostic aspects, disease activity and response to therapy, so-called biomarkers Increasing digitalization and the availability of easy-to-use devices and technology also enable healthcare professionals to use a new class of digital biomarkers—digital health technologies—to explain, influence and/or predict health-related outcomes. The technology and devices from which these digital biomarkers stem are quite broad, and range from wearables that collect patients’ activity during digitalized functional tests (e.g., the Multiple Sclerosis Performance Test, dual-tasking performance and speech) to digitalized diagnostic procedures (e.g., optical coherence tomography) and software-supported magnetic resonance imaging evaluation. These technologies offer a timesaving way to collect valuable data on a regular basis over a long period of time, not only once or twice a year during patients’ routine visit at the clinic. Therefore, they lead to real-life data acquisition, closer patient monitoring and thus a patient dataset useful for precision medicine. Despite the great benefit of such increasing digitalization, for now, the path to implementing digital biomarkers is widely unknown or inconsistent. Challenges around validation, infrastructure, evidence generation, consistent data collection and analysis still persist. In this narrative review, we explore existing and future opportunities to capture clinical digital biomarkers in the care of people with MS, which may lead to a digital twin of the patient. To do this, we searched published papers for existing opportunities to capture clinical digital biomarkers for different functional systems in the context of MS, and also gathered perspectives on digital biomarkers under development or already existing as a research approach.

show abstract

Vocal signal analysis in patients affected by Multiple Sclerosis

Cited by 20 publications

References 19 publications

Optimising Speaker-Dependent Feature Extraction Parameters to Improve Automatic Speech Recognition Performance for People with Dysarthria

Optimising Speaker-Dependent Feature Extraction Parameters to Improve Automatic Speech Recognition Performance for People with Dysarthria

On the Use of Voice Signals for Studying Sclerosis Disease

Digital Biomarkers in Multiple Sclerosis

Contact Info

Product

Resources

About