An AI-Driven Predictive Modelling Framework to Analyze and Visualize Blood Product Transactional Data for Reducing Blood Products’ Discards

Jaber Rad, Calvino Cheng, Jason G. Quinn, Samina Abidi, Robert Liwski, Syed Sibte Raza Abidi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Maintaining an equilibrium between shortage and wastage in blood inventories is challenging due to the perishable nature of blood products. Research in blood product inventory management has predominantly focused on reducing wastage due to outdates (i.e. expiry of the blood product), whereas wastage due to discards, related to the lifecycle of a blood product, is not well investigated. In this study, we investigate machine learning methods to analyze blood product transition sequences in a large real-life transactional dataset of Red Blood Cells (RBC) to predict potential blood product discard. Our prediction models are able to predict with 79% accuracy potential discards based on the blood product’s current transaction data. We applied advanced data visualizations methods to develop an interactive blood inventory dashboard to help laboratory managers to probe blood units’ lifecycles to identify discard causes.

Original languageEnglish
Title of host publicationArtificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings
EditorsMartin Michalowski, Robert Moskovitch
PublisherSpringer Science and Business Media Deutschland GmbH
Pages192-202
Number of pages11
ISBN (Print)9783030591366
DOIs
Publication statusPublished - 2020
Event18th International Conference on Artificial Intelligence in Medicine, AIME 2020 - Minneapolis, United States
Duration: Aug 25 2020Aug 28 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12299 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th International Conference on Artificial Intelligence in Medicine, AIME 2020
Country/TerritoryUnited States
CityMinneapolis
Period8/25/208/28/20

Bibliographical note

Funding Information:
This research is supported by the Blood Efficiency Accelerator Award by Canadian Blood Services. We thank the NSHA Central Zone Blood Transfusion Services for providing us the dataset and supporting the project.

Funding Information:
Acknowledgements. This research is supported by the Blood Efficiency Accelerator Award by Canadian Blood Services. We thank the NSHA Central Zone Blood Transfusion Services for providing us the dataset and supporting the project.

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

ASJC Scopus Subject Areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'An AI-Driven Predictive Modelling Framework to Analyze and Visualize Blood Product Transactional Data for Reducing Blood Products’ Discards'. Together they form a unique fingerprint.

Cite this