Pareto-Optimized AVQI Assessment of Dysphonia: A Clinical Trial Using Various Smartphones

Multiparametric indices offer a more comprehensive approach to voice quality assessment by taking into account multiple acoustic parameters. Artificial intelligence technology can be utilized in healthcare evaluate data and optimize decision-making processes. Mobile devices provide new opportunities for remote speech monitoring, allowing the use of basic mobile as screening tools early identification treatment disorders. However, it is necessary demonstrate equivalence between device signals gold standard microphone preamplifiers. Despite increased availability technology, there still lack understanding impact physiological, speech/language, cultural factors on assessment. Challenges research include accounting organic speech-related covariables, such differences conversing sound pressure level (SPL) fundamental frequency (f0), recognizing link sensory experimental outcomes, obtaining large dataset understand regular variation within voice-disordered individuals. Our study investigated cellphones estimate Acoustic Voice Quality Index (AVQI) typical clinical setting using Pareto-optimized signal processing path. We found that was strong correlation AVQI results obtained from different smartphones studio microphone, with no significant mean scores smartphones. The diagnostic accuracy comparable professional optimal cut-off values effectively distinguish normal pathological each smartphone used study. All met proposed 0.8 AUC threshold demonstrated an acceptable Youden index value.