EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

The effect of resampling techniques on the performances of machine learning clinical risk prediction models in the setting of severe class imbalance: development and internal validation in a retrospective cohort

Janny Xue Chen Ke iD, Arunachalam DhakshinaMurthy, Ronald B George, Paula Branco

Discover Artificial Intelligence · 26.11.2024 · 4 (1)

DOI: 10.1007/s44163-024-00199-0 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Discover Artificial Intelligence, online erschienen am 26.11.2024. Verfasst von Janny Xue Chen Ke und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Anesthesia, St. Paul's Hospital, Providence Health Care, 1081 Burrard Street, Vancouver, BC V6Z1Y6 Canada. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Nova Scotia Health Authority, Natural Sciences and Engineering Research Council of Canada.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

The availability of population datasets and machine learning techniques heralded a new era of sophisticated prediction models involving a large number of routinely collected variables. However, severe class imbalance in clinical datasets is a major challenge. The aim of this study is to investigate the impact of commonly-used resampling techniques in combination with commonly-used machine learning algorithms in a clinical dataset, to determine whether combination(s) of these approaches improve upon the original multivariable logistic regression with no resampling.

Methods

We previously developed and internally validated a multivariable logistic regression 30-day mortality prediction model in 30,619 patients using preoperative and intraoperative features.Using the same dataset, we systematically evaluated and compared model performances after application of resampling techniques [random under-sampling, near miss under-sampling, random oversampling, and synthetic minority oversampling (SMOTE)] in combination with machine learning algorithms (logistic regression, elastic net, decision trees, random forest, and extreme gradient boosting).

Results

We found that in the setting of severe class imbalance, the impact of resampling techniques on model performance varied by the machine learning algorithm and the evaluation metric. Existing resampling techniques did not meaningfully improve area under receiving operating curve (AUROC). The area under the precision recall curve (AUPRC) was only increased by random under-sampling and SMOTE for decision trees, and oversampling and SMOTE for extreme gradient boosting. Importantly, some combinations of algorithm and resampling technique decreased AUROC and AUPRC compared to no resampling.

Conclusion

Existing resampling techniques had a variable impact on models, depending on the algorithms and the evaluation metrics. Future research is needed to improve predictive performances in the setting of severe class imbalance.

Schlagwörter

Stichwörter der Autor:innen

AnesthesiologyClass imbalanceMachine learningPredictive modelingResamplingRisk prediction

Autor:innen und Institutionen

  • Department of Anesthesia, St. Paul's Hospital, Providence Health Care, 1081 Burrard Street, Vancouver, BC V6Z1Y6 Canada.
  • Department of Anesthesiology, Pharmacology and Therapeutics, University of British Columbia, Vancouver, BC Canada.
  • Department of Anesthesia, Pain Management & Perioperative Medicine, Dalhousie University, Halifax, NS Canada.
  • School of Computer Science, Carleton University, Ottawa, ON Canada.
  • Mount Sinai Hospital, University of Toronto, Toronto, ON Canada.
  • School of Electrical Engineering and Computer Science, University of Ottawa, Ottawa, ON Canada.

ORCID-Profile: Janny Xue Chen Ke

Förderung

  • Nova Scotia Health Authority
  • Natural Sciences and Engineering Research Council of Canada

Zitieren

APA

Ke, J. X. C., DhakshinaMurthy, A., George, R. B., & Branco, P. (2024). The effect of resampling techniques on the performances of machine learning clinical risk prediction models in the setting of severe class imbalance: development and internal validation in a retrospective cohort. Discover Artificial Intelligence, 4(1). https://doi.org/10.1007/s44163-024-00199-0

Vancouver

Ke JXC, DhakshinaMurthy A, George RB, Branco P. The effect of resampling techniques on the performances of machine learning clinical risk prediction models in the setting of severe class imbalance: development and internal validation in a retrospective cohort. Discover Artificial Intelligence. 2024;4(1). doi:10.1007/s44163-024-00199-0

BibTeX
@article{Ke2024,
  author = {Ke, Janny Xue Chen and DhakshinaMurthy, Arunachalam and George, Ronald B and Branco, Paula},
  title = {The effect of resampling techniques on the performances of machine learning clinical risk prediction models in the setting of severe class imbalance: development and internal validation in a retrospective cohort},
  journal = {Discover Artificial Intelligence},
  year = {2024},
  volume = {4},
  number = {1},
  doi = {10.1007/s44163-024-00199-0}
}
Kostenlos starten →