Replication-based regularization approaches to diagnose Reinke's edema by using voice recordings

作者：

Highlights：

• Development of Bayesian regularization approaches based on replicated measurements

• Capturing within-subject variability versus approaches based on independent instances

• Experiment conducted to detect Reinke's edema

• Speech features based on nonlinear methods and source-filter theory

• Robust performance in terms of feature selection and predictive capability

摘要

•Development of Bayesian regularization approaches based on replicated measurements•Capturing within-subject variability versus approaches based on independent instances•Experiment conducted to detect Reinke's edema•Speech features based on nonlinear methods and source-filter theory•Robust performance in terms of feature selection and predictive capability

论文关键词：Acoustic features,Classification,Reinke's edema,Regularization,Replicated measurements,Variable selection

论文评审过程：Received 15 February 2021, Revised 21 August 2021, Accepted 31 August 2021, Available online 8 September 2021, Version of Record 10 September 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102162