About Me
About you ..
Research Interests
Your research interests.
Publications
2020
Jaber Rad; Calvino Cheng; Jason G Quinn; Samina Abidi; Robert Liwski; Syed Sibte Raza Abidi
An AI-Driven Predictive Modelling Framework to Analyze and Visualize Blood Product Transactional Data for Reducing Blood Products’ Discards Proceedings Article
In: International Conference on Artificial Intelligence in Medicine (AIME 2020), 2020.
Abstract | BibTeX | Tags: Big Data, Blood Inventory Management, Blood product wastage, Data visualization, Machine Learning, Sequence prediction, Visual analytics
@inproceedings{JRAD2020,
title = {An AI-Driven Predictive Modelling Framework to Analyze and Visualize Blood Product Transactional Data for Reducing Blood Products’ Discards},
author = {Jaber Rad and Calvino Cheng and Jason G Quinn and Samina Abidi and Robert Liwski and Syed Sibte Raza Abidi},
year = {2020},
date = {2020-08-01},
urldate = {2020-08-01},
booktitle = {International Conference on Artificial Intelligence in Medicine (AIME 2020)},
abstract = {Maintaining an equilibrium between shortage and wastage in blood inventories is challenging due to the perishable nature of blood products. Re-search in blood product inventory management has predominantly focused on reducing wastage due to outdates (i.e. expiry of the blood product), whereas wast-age 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.},
keywords = {Big Data, Blood Inventory Management, Blood product wastage, Data visualization, Machine Learning, Sequence prediction, Visual analytics},
pubstate = {published},
tppubtype = {inproceedings}
}
Maintaining an equilibrium between shortage and wastage in blood inventories is challenging due to the perishable nature of blood products. Re-search in blood product inventory management has predominantly focused on reducing wastage due to outdates (i.e. expiry of the blood product), whereas wast-age 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.