EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Demand forecasting for platelet usage: From univariate time series to multivariable models

Maryam Motamedi, Jessica Dawson, Na Li, Douglas G Down iD, Nancy M Heddle

PLOS ONE · 23.04.2024 · 19 (4) · e0297391

DOI: 10.1371/journal.pone.0297391 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 23.04.2024. Verfasst von Maryam Motamedi und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computing and Software, McMaster University, Hamilton, Ontario, Canada. Hauptthemen laut MEDLINE: Platelet Transfusion, Forecasting. Die Arbeit wurde bisher 28-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NSERC Discovery Grant program and Mitacs through the Accelerate Industrial Postdoc program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Platelet products are both expensive and have very short shelf lives. As usage rates for platelets are highly variable, the effective management of platelet demand and supply is very important yet challenging. The primary goal of this paper is to present an efficient forecasting model for platelet demand at Canadian Blood Services (CBS). To accomplish this goal, five different demand forecasting methods, ARIMA (Auto Regressive Integrated Moving Average), Prophet, lasso regression (least absolute shrinkage and selection operator), random forest, and LSTM (Long Short-Term Memory) networks are utilized and evaluated via a rolling window method. We use a large clinical dataset for a centralized blood distribution centre for four hospitals in Hamilton, Ontario, spanning from 2010 to 2018 and consisting of daily platelet transfusions along with information such as the product specifications, the recipients’ characteristics, and the recipients’ laboratory test results. This study is the first to utilize different methods from statistical time series models to data-driven regression and machine learning techniques for platelet transfusion using clinical predictors and with different amounts of data. We find that the multivariable approaches have the highest accuracy in general, however, if sufficient data are available, a simpler time series approach appears to be sufficient. We also comment on the approach to choose predictors for the multivariable models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Platelet TransfusionForecastingHumansBlood PlateletsMaleFemaleOntarioMachine LearningMiddle AgedModels, StatisticalAgedMultivariate Analysis

Autor:innen und Institutionen

  • Department of Computing and Software, McMaster University, Hamilton, Ontario, Canada.
  • Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada.
  • Michael G. DeGroote Centre for Transfusion Research, Faculty of Health Sciences, Hamilton, Ontario, Canada.
  • Centre for Innovation, Canadian Blood Services, Ottawa, Ontario, Canada.

ORCID-Profile: Douglas G Down

Förderung

  • NSERC Discovery Grant program and Mitacs through the Accelerate Industrial Postdoc program

Zitieren

APA

Motamedi, M., Dawson, J., Li, N., Down, D. G., & Heddle, N. M. (2024). Demand forecasting for platelet usage: From univariate time series to multivariable models. PLOS ONE, 19(4), e0297391. https://doi.org/10.1371/journal.pone.0297391

Vancouver

Motamedi M, Dawson J, Li N, Down DG, Heddle NM. Demand forecasting for platelet usage: From univariate time series to multivariable models. PLOS ONE. 2024;19(4):e0297391. doi:10.1371/journal.pone.0297391

BibTeX
@article{Motamedi2024,
  author = {Motamedi, Maryam and Dawson, Jessica and Li, Na and Down, Douglas G and Heddle, Nancy M},
  title = {Demand forecasting for platelet usage: From univariate time series to multivariable models},
  journal = {PLOS ONE},
  year = {2024},
  volume = {19},
  number = {4},
  pages = {e0297391},
  doi = {10.1371/journal.pone.0297391}
}
Kostenlos starten →