Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features

研究成果: 著書/レポートタイプへの貢献会議での発言

抄録

Estimating individual causal effect is important for decision making in many fields especially for medical interventions. We propose an interpretable and accurate algorithm for estimating causal effects from observational data. The proposed scheme is combining multiple predictors' outputs by an interpretable predictor such as linear predictor and if then rules. We secure interpretability using the interpretable predictor and balancing scores in causal inference studies as meta-features. For securing accuracy, we adapt machine learning algorithms for calculating balancing scores. We analyze the effect of t-PA therapy for stroke patients using real-world data, which has 64,609 records with 362 variables and interpret results. The results show that cross validation AUC of the proposed scheme is little less than original machine learning scheme; however, the proposed scheme provides interpretability that t-PA therapy is effective for severe patients.

元の言語英語
ホスト出版物のタイトル40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
出版者Institute of Electrical and Electronics Engineers Inc.
ページ4042-4045
ページ数4
ISBN(電子版)9781538636466
DOI
出版物ステータス出版済み - 10 26 2018
イベント40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018 - Honolulu, 米国
継続期間: 7 18 20187 21 2018

出版物シリーズ

名前Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
2018-July
ISSN(印刷物)1557-170X

その他

その他40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
米国
Honolulu
期間7/18/187/21/18

Fingerprint

Learning systems
Learning algorithms
Area Under Curve
Decision Making
Decision making
Stroke
Therapeutics
Machine Learning

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

これを引用

Nohara, Y., Iihara, K., & Nakashima, N. (2018). Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features. : 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018 (pp. 4042-4045). [8513026] (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS; 巻数 2018-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/EMBC.2018.8513026

Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features. / Nohara, Yasunobu; Iihara, Koji; Nakashima, Naoki.

40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 4042-4045 8513026 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS; 巻 2018-July).

研究成果: 著書/レポートタイプへの貢献会議での発言

Nohara, Y, Iihara, K & Nakashima, N 2018, Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features. : 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018., 8513026, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, 巻. 2018-July, Institute of Electrical and Electronics Engineers Inc., pp. 4042-4045, 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018, Honolulu, 米国, 7/18/18. https://doi.org/10.1109/EMBC.2018.8513026
Nohara Y, Iihara K, Nakashima N. Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features. : 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 4042-4045. 8513026. (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS). https://doi.org/10.1109/EMBC.2018.8513026
Nohara, Yasunobu ; Iihara, Koji ; Nakashima, Naoki. / Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features. 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 4042-4045 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS).
@inproceedings{08725b5227ce44328c543de15db2a15f,
title = "Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features",
abstract = "Estimating individual causal effect is important for decision making in many fields especially for medical interventions. We propose an interpretable and accurate algorithm for estimating causal effects from observational data. The proposed scheme is combining multiple predictors' outputs by an interpretable predictor such as linear predictor and if then rules. We secure interpretability using the interpretable predictor and balancing scores in causal inference studies as meta-features. For securing accuracy, we adapt machine learning algorithms for calculating balancing scores. We analyze the effect of t-PA therapy for stroke patients using real-world data, which has 64,609 records with 362 variables and interpret results. The results show that cross validation AUC of the proposed scheme is little less than original machine learning scheme; however, the proposed scheme provides interpretability that t-PA therapy is effective for severe patients.",
author = "Yasunobu Nohara and Koji Iihara and Naoki Nakashima",
year = "2018",
month = "10",
day = "26",
doi = "10.1109/EMBC.2018.8513026",
language = "English",
series = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "4042--4045",
booktitle = "40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018",
address = "United States",

}

TY - GEN

T1 - Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features

AU - Nohara, Yasunobu

AU - Iihara, Koji

AU - Nakashima, Naoki

PY - 2018/10/26

Y1 - 2018/10/26

N2 - Estimating individual causal effect is important for decision making in many fields especially for medical interventions. We propose an interpretable and accurate algorithm for estimating causal effects from observational data. The proposed scheme is combining multiple predictors' outputs by an interpretable predictor such as linear predictor and if then rules. We secure interpretability using the interpretable predictor and balancing scores in causal inference studies as meta-features. For securing accuracy, we adapt machine learning algorithms for calculating balancing scores. We analyze the effect of t-PA therapy for stroke patients using real-world data, which has 64,609 records with 362 variables and interpret results. The results show that cross validation AUC of the proposed scheme is little less than original machine learning scheme; however, the proposed scheme provides interpretability that t-PA therapy is effective for severe patients.

AB - Estimating individual causal effect is important for decision making in many fields especially for medical interventions. We propose an interpretable and accurate algorithm for estimating causal effects from observational data. The proposed scheme is combining multiple predictors' outputs by an interpretable predictor such as linear predictor and if then rules. We secure interpretability using the interpretable predictor and balancing scores in causal inference studies as meta-features. For securing accuracy, we adapt machine learning algorithms for calculating balancing scores. We analyze the effect of t-PA therapy for stroke patients using real-world data, which has 64,609 records with 362 variables and interpret results. The results show that cross validation AUC of the proposed scheme is little less than original machine learning scheme; however, the proposed scheme provides interpretability that t-PA therapy is effective for severe patients.

UR - http://www.scopus.com/inward/record.url?scp=85056635136&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056635136&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2018.8513026

DO - 10.1109/EMBC.2018.8513026

M3 - Conference contribution

T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

SP - 4042

EP - 4045

BT - 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -