%0 Journal Article %1 voigt2011 %A Voigt, D. %A Eysholdt, U. %D 2011 %J J Acoust Soc Am %K mda %N 4 %P 2550 %T Identifying relevant analysis parameters for the classification of vocal fold dynamics. %V 130 %X In the previous work, a computer-based analysis framework was proposed, which is capable of objectively and automatically classifying vocal fold vibrations as captured by high-speed videoendoscopy during phonation. The method is based on quantitative feature extraction from Phonovibrograms combined with nonlinear machine learning techniques, allowing for the discrimination of normal and pathological laryngeal movement patterns. The diagnostic reliability and potential of this analysis approach were demonstrated. However, the practically relevant question, whether certain control parameters of the procedure can lead to increased classification accuracy, remained partially unanswered. In this study, the following parameter sets of the analysis framework were investigated in a systematic manner: method of feature extraction, type of feature aggregation and normalization, number of considered oscillation cycles, feature laterality, classification task, and employed machine learning algorithm. For this purpose, more than 150 000 experiments were conducted using a data set of 105 laryngeal high-speed video recordings, comprising various clinical cases with non-organic findings and subjects from a healthy control group. The results of this extensive study show the particular suitability of certain parameter combinations, helping to further improve the practical application of the automated classification framework for vocal fold dynamics.

@article{voigt2011, abstract = {In the previous work, a computer-based analysis framework was proposed, which is capable of objectively and automatically classifying vocal fold vibrations as captured by high-speed videoendoscopy during phonation. The method is based on quantitative feature extraction from Phonovibrograms combined with nonlinear machine learning techniques, allowing for the discrimination of normal and pathological laryngeal movement patterns. The diagnostic reliability and potential of this analysis approach were demonstrated. However, the practically relevant question, whether certain control parameters of the procedure can lead to increased classification accuracy, remained partially unanswered. In this study, the following parameter sets of the analysis framework were investigated in a systematic manner: method of feature extraction, type of feature aggregation and normalization, number of considered oscillation cycles, feature laterality, classification task, and employed machine learning algorithm. For this purpose, more than 150 000 experiments were conducted using a data set of 105 laryngeal high-speed video recordings, comprising various clinical cases with non-organic findings and subjects from a healthy control group. The results of this extensive study show the particular suitability of certain parameter combinations, helping to further improve the practical application of the automated classification framework for vocal fold dynamics. }, added-at = {2012-02-15T13:32:12.000+0100}, author = {Voigt, D. and Eysholdt, U.}, biburl = {https://www.bibsonomy.org/bibtex/22cee1fd0755167c047b5ada9d7b1502d/nschlitt}, interhash = {84a8a9fc0f2d060766c646dfbd761ed9}, intrahash = {2cee1fd0755167c047b5ada9d7b1502d}, journal = {J Acoust Soc Am}, keywords = {mda}, number = 4, pages = 2550, timestamp = {2012-02-15T13:32:12.000+0100}, title = {Identifying relevant analysis parameters for the classification of vocal fold dynamics.}, volume = 130, year = 2011 }