This paper focuses on Cepstral Peak Prominence Smoothed (CPPS) as a possible indicator of vocal health status, considering individual CPPS distribution and its descriptive statistics. 31 voluntary patients and 22 control subjects performed the same protocol, which includes the simultaneous acquisition of three repetitions of the sustained vowel /a/ with a microphone in air and a contact sensor, the perceptual assessment of voice and the videolaringoscopy examination. The best logistic regression models have been applied and preliminary results showed that the fifth percentile and the standard deviation of CPPS distributions are the best parameters that discriminate healthy and unhealthy voice for the microphone in air and the contact sensor, respectively. The Area Under Curve (AUC) revealed the diagnostic precision of the selected CPPS parameters: AUC of 0.96 and 0.83 have been found for the microphone in air and the contact sensor, showing strong to moderate discrimination power, respectively. The repeatability of the selected CPPS parameters has been also estimated. For each selected CPPS parameter, the Monte Carlo method has been implemented in order to evaluate the uncertainty of the threshold, which was identified by means of the Receiver Operating Curve analysis.

Cepstral Peak Prominence Smoothed distribution as discriminator of vocal health in sustained vowel / Castellana, Antonella; Carullo, Alessio; Corbellini, Simone; Astolfi, Arianna; M., Spadola Bisetti; J., Colombini. - ELETTRONICO. - Unico:(2017), pp. 552-557. ((Intervento presentato al convegno 2017 IEEE International Instrumentation and Measurement Technology Conference tenutosi a Torino nel May, 22-25.

Cepstral Peak Prominence Smoothed distribution as discriminator of vocal health in sustained vowel

CASTELLANA, ANTONELLA;CARULLO, Alessio;CORBELLINI, SIMONE;ASTOLFI, Arianna;
2017

Abstract

This paper focuses on Cepstral Peak Prominence Smoothed (CPPS) as a possible indicator of vocal health status, considering individual CPPS distribution and its descriptive statistics. 31 voluntary patients and 22 control subjects performed the same protocol, which includes the simultaneous acquisition of three repetitions of the sustained vowel /a/ with a microphone in air and a contact sensor, the perceptual assessment of voice and the videolaringoscopy examination. The best logistic regression models have been applied and preliminary results showed that the fifth percentile and the standard deviation of CPPS distributions are the best parameters that discriminate healthy and unhealthy voice for the microphone in air and the contact sensor, respectively. The Area Under Curve (AUC) revealed the diagnostic precision of the selected CPPS parameters: AUC of 0.96 and 0.83 have been found for the microphone in air and the contact sensor, showing strong to moderate discrimination power, respectively. The repeatability of the selected CPPS parameters has been also estimated. For each selected CPPS parameter, the Monte Carlo method has been implemented in order to evaluate the uncertainty of the threshold, which was identified by means of the Receiver Operating Curve analysis.
978-1-5090-3596-0
File in questo prodotto:
File Dimensione Formato  
Castellana et al_review_nero.pdf

accesso aperto

Descrizione: Versione finale inviata al congresso
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 244.93 kB
Formato Adobe PDF
244.93 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11583/2675359
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo