Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques

<i>Background and Objectives</i>: Determining the presence or absence of cochlear dead regions (DRs) is essential in clinical practice. This study proposes a machine learning (ML)-based model that applies oversampling techniques for predicting DRs in patients. <i>Materials and Meth...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Young-Soo Chang, Hee-Sung Park, Il-Joon Moon
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	cochlear dead region machine learning prediction model oversampling method synthetic minority oversampling technique Medicine (General) R5-920
Acceso en línea:	https://doaj.org/article/1caa348bdf074d1fb840d4a2927862d5
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1caa348bdf074d1fb840d4a2927862d5
record_format	dspace
spelling	oai:doaj.org-article:1caa348bdf074d1fb840d4a2927862d52021-11-25T18:18:28ZPredicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques10.3390/medicina571111921648-91441010-660Xhttps://doaj.org/article/1caa348bdf074d1fb840d4a2927862d52021-11-01T00:00:00Zhttps://www.mdpi.com/1648-9144/57/11/1192https://doaj.org/toc/1010-660Xhttps://doaj.org/toc/1648-9144<i>Background and Objectives</i>: Determining the presence or absence of cochlear dead regions (DRs) is essential in clinical practice. This study proposes a machine learning (ML)-based model that applies oversampling techniques for predicting DRs in patients. <i>Materials and Methods</i>: We used recursive partitioning and regression for classification tree (CT) and logistic regression (LR) as prediction models. To overcome the imbalanced nature of the dataset, oversampling techniques to duplicate examples in the minority class or to synthesize new examples from existing examples in the minority class were adopted, namely the synthetic minority oversampling technique (SMOTE). <i>Results</i>: The accuracy results of the 10-fold cross-validation of the LR and CT with the original data were 0.82 (±0.02) and 0.93 (±0.01), respectively. The accuracy results of the 10-fold cross-validation of the LR and CT with the oversampled data were 0.66 (±0.02) and 0.86 (±0.01), respectively. <i>Conclusions</i>: This study is the first to adopt the SMOTE method to assess the role of oversampling methods on audiological datasets and to develop an ML-based model. Considering that the SMOTE method did not improve the model’s performance, a more flexible model or more clinical features may be needed.Young-Soo ChangHee-Sung ParkIl-Joon MoonMDPI AGarticlecochlear dead regionmachine learningprediction modeloversampling methodsynthetic minority oversampling techniqueMedicine (General)R5-920ENMedicina, Vol 57, Iss 1192, p 1192 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	cochlear dead region machine learning prediction model oversampling method synthetic minority oversampling technique Medicine (General) R5-920
spellingShingle	cochlear dead region machine learning prediction model oversampling method synthetic minority oversampling technique Medicine (General) R5-920 Young-Soo Chang Hee-Sung Park Il-Joon Moon Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques
description	<i>Background and Objectives</i>: Determining the presence or absence of cochlear dead regions (DRs) is essential in clinical practice. This study proposes a machine learning (ML)-based model that applies oversampling techniques for predicting DRs in patients. <i>Materials and Methods</i>: We used recursive partitioning and regression for classification tree (CT) and logistic regression (LR) as prediction models. To overcome the imbalanced nature of the dataset, oversampling techniques to duplicate examples in the minority class or to synthesize new examples from existing examples in the minority class were adopted, namely the synthetic minority oversampling technique (SMOTE). <i>Results</i>: The accuracy results of the 10-fold cross-validation of the LR and CT with the original data were 0.82 (±0.02) and 0.93 (±0.01), respectively. The accuracy results of the 10-fold cross-validation of the LR and CT with the oversampled data were 0.66 (±0.02) and 0.86 (±0.01), respectively. <i>Conclusions</i>: This study is the first to adopt the SMOTE method to assess the role of oversampling methods on audiological datasets and to develop an ML-based model. Considering that the SMOTE method did not improve the model’s performance, a more flexible model or more clinical features may be needed.
format	article
author	Young-Soo Chang Hee-Sung Park Il-Joon Moon
author_facet	Young-Soo Chang Hee-Sung Park Il-Joon Moon
author_sort	Young-Soo Chang
title	Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques
title_short	Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques
title_full	Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques
title_fullStr	Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques
title_full_unstemmed	Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques
title_sort	predicting the cochlear dead regions using a machine learning-based approach with oversampling techniques
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/1caa348bdf074d1fb840d4a2927862d5
work_keys_str_mv	AT youngsoochang predictingthecochleardeadregionsusingamachinelearningbasedapproachwithoversamplingtechniques AT heesungpark predictingthecochleardeadregionsusingamachinelearningbasedapproachwithoversamplingtechniques AT iljoonmoon predictingthecochleardeadregionsusingamachinelearningbasedapproachwithoversamplingtechniques
_version_	1718411367245938688

Predicting the Cochlear Dead Regions Using a Machine Learning-Based Approach with Oversampling Techniques

Ejemplares similares