The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning

Hsiang Wei Hu; Jiun Yi Yang; Chi Hin Un; Kuan Yu Chen; Chih Chiang Huang; Rua Huan Tsaih; Chou Ching K. Lin; Hsuan Chang; Hsuan Ming Lin

doi:10.1109/ECBIOS51820.2021.9510559

The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning

Hsiang Wei Hu, Jiun Yi Yang, Chi Hin Un, Kuan Yu Chen, Chih Chiang Huang, Rua Huan Tsaih, Chou Ching K. Lin, Hsuan Chang, Hsuan Ming Lin

Department of Neurology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Citations (Scopus)

Abstract

Intradialytic hypotension (IDH) is a common complication and make patient get brain and heart disease. This study aimed to use the new method of feature selection to improve the performance of IDH prediction. We used feature importance by a different model to sort from high to low one by one, remove the lowest feature importance, and follow the rule again. Therefore, we obtain the combination of feature values that have the highest accuracy for predictive power. After applying REFCV with linear regression, the AUC of ROC increased to 0.966 (2-layers LSTM) and 0.965 (3-layers LSTM). The feature selection method we designed resulted in a better prediction ability in the model.

Original language	English
Title of host publication	3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021
Editors	Teen-Hang Meen
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	69-70
Number of pages	2
ISBN (Electronic)	9781728193045
DOIs	https://doi.org/10.1109/ECBIOS51820.2021.9510559
Publication status	Published - 2021
Event	3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021 - Tainan, Taiwan Duration: 2021 May 28 → 2021 May 30

Publication series

Name	3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021

Conference

Conference	3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021
Country/Territory	Taiwan
City	Tainan
Period	21-05-28 → 21-05-30

All Science Journal Classification (ASJC) codes

Strategy and Management
Renewable Energy, Sustainability and the Environment
Biomedical Engineering
Safety, Risk, Reliability and Quality
Building and Construction
Health Policy
Health(social science)

Access to Document

10.1109/ECBIOS51820.2021.9510559

Cite this

Hu, H. W., Yang, J. Y., Un, C. H., Chen, K. Y., Huang, C. C., Tsaih, R. H., Lin, C. C. K., Chang, H., & Lin, H. M. (2021). The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning. In T.-H. Meen (Ed.), 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021 (pp. 69-70). (3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ECBIOS51820.2021.9510559

Hu, Hsiang Wei ; Yang, Jiun Yi ; Un, Chi Hin et al. / The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning. 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021. editor / Teen-Hang Meen. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 69-70 (3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021).

@inproceedings{0bd29c5d3a0b4e37bd2fe283fbc5e675,

title = "The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning",

abstract = "Intradialytic hypotension (IDH) is a common complication and make patient get brain and heart disease. This study aimed to use the new method of feature selection to improve the performance of IDH prediction. We used feature importance by a different model to sort from high to low one by one, remove the lowest feature importance, and follow the rule again. Therefore, we obtain the combination of feature values that have the highest accuracy for predictive power. After applying REFCV with linear regression, the AUC of ROC increased to 0.966 (2-layers LSTM) and 0.965 (3-layers LSTM). The feature selection method we designed resulted in a better prediction ability in the model.",

author = "Hu, {Hsiang Wei} and Yang, {Jiun Yi} and Un, {Chi Hin} and Chen, {Kuan Yu} and Huang, {Chih Chiang} and Tsaih, {Rua Huan} and Lin, {Chou Ching K.} and Hsuan Chang and Lin, {Hsuan Ming}",

note = "Publisher Copyright: {\textcopyright} 2021 ECBIOS 2021. All rights reserved.; 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021 ; Conference date: 28-05-2021 Through 30-05-2021",

year = "2021",

doi = "10.1109/ECBIOS51820.2021.9510559",

language = "English",

series = "3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "69--70",

editor = "Teen-Hang Meen",

booktitle = "3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021",

address = "United States",

}

Hu, HW, Yang, JY, Un, CH, Chen, KY, Huang, CC, Tsaih, RH, Lin, CCK, Chang, H & Lin, HM 2021, The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning. in T-H Meen (ed.), 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021. 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021, Institute of Electrical and Electronics Engineers Inc., pp. 69-70, 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021, Tainan, Taiwan, 21-05-28. https://doi.org/10.1109/ECBIOS51820.2021.9510559

The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning. / Hu, Hsiang Wei; Yang, Jiun Yi; Un, Chi Hin et al.
3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021. ed. / Teen-Hang Meen. Institute of Electrical and Electronics Engineers Inc., 2021. p. 69-70 (3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning

AU - Hu, Hsiang Wei

AU - Yang, Jiun Yi

AU - Un, Chi Hin

AU - Chen, Kuan Yu

AU - Huang, Chih Chiang

AU - Tsaih, Rua Huan

AU - Lin, Chou Ching K.

AU - Chang, Hsuan

AU - Lin, Hsuan Ming

PY - 2021

Y1 - 2021

N2 - Intradialytic hypotension (IDH) is a common complication and make patient get brain and heart disease. This study aimed to use the new method of feature selection to improve the performance of IDH prediction. We used feature importance by a different model to sort from high to low one by one, remove the lowest feature importance, and follow the rule again. Therefore, we obtain the combination of feature values that have the highest accuracy for predictive power. After applying REFCV with linear regression, the AUC of ROC increased to 0.966 (2-layers LSTM) and 0.965 (3-layers LSTM). The feature selection method we designed resulted in a better prediction ability in the model.

AB - Intradialytic hypotension (IDH) is a common complication and make patient get brain and heart disease. This study aimed to use the new method of feature selection to improve the performance of IDH prediction. We used feature importance by a different model to sort from high to low one by one, remove the lowest feature importance, and follow the rule again. Therefore, we obtain the combination of feature values that have the highest accuracy for predictive power. After applying REFCV with linear regression, the AUC of ROC increased to 0.966 (2-layers LSTM) and 0.965 (3-layers LSTM). The feature selection method we designed resulted in a better prediction ability in the model.

UR - http://www.scopus.com/inward/record.url?scp=85124872842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85124872842&partnerID=8YFLogxK

U2 - 10.1109/ECBIOS51820.2021.9510559

DO - 10.1109/ECBIOS51820.2021.9510559

M3 - Conference contribution

AN - SCOPUS:85124872842

T3 - 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021

SP - 69

EP - 70

BT - 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021

A2 - Meen, Teen-Hang

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021

Y2 - 28 May 2021 through 30 May 2021

ER -

Hu HW, Yang JY, Un CH, Chen KY, Huang CC, Tsaih RH et al. The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning. In Meen TH, editor, 3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 69-70. (3rd IEEE Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2021). doi: 10.1109/ECBIOS51820.2021.9510559

The New Method of Feature Selection for Intradialytic Hypotension Prediction using Machine learning

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this