Referral prediction in Healthcare using Graph Neural Networks
Published in GReS – Workshop on Graph Neural Networks for Recommendation and Search. Co-located with the ACM RecSys ’21 conference, 2021
Recommended citation: REGINA DUARTE, QIWEI HAN, CLAUDIA SOARES, "Referral prediction in Healthcare using Graph Neural Networks." GReS – Workshop on Graph Neural Networks for Recommendation and Search. Co-located with the ACM RecSys ’21 conference, 2021. https://europe.naverlabs.com/wp-content/uploads/2021/09/DuarteEtAl2021.pdf
Abstract
Medical referrals between primary care doctors and specialists affect many aspects of patient care, such as quality of care, patient satisfaction, and health care costs. In this work, we hypothesize that the social network of medical doctors can influence the referral process. We analyse primary-specialty referrals through transaction data over medical appointments gathered between 2012 and 2017 in all hospital centers of an European private healthcare provider. In this paper we uncover patterns and hidden mechanisms in primary-specialty referrals to improve the efficiency of health care services, and propose a link prediction model for recommendation using Graph Neural Networks. First, we carried out exploratory data analysis and searched for patterns or interesting features of the data and afterwords, we created and analysed the doctor’s social network and the referrals network. We learned the representation of the referral network using a Graph Neural Network (GNN). Finally, we discussed the nodes representation of the referral network which evidences our main hypothesis. This work addresses the discovery of important patterns in medical referrals to, in future work, improve the efficiency of collaboration within organizations.
BibTeX
@article{duarte2021referral,
author = "DUARTE, REGINA and HAN, QIWEI and SOARES, CLAUDIA",
title = "Referral prediction in Healthcare using Graph Neural Networks",
year = "2021",
journal = "GReS – Workshop on Graph Neural Networks for Recommendation and Search. Co-located with the ACM RecSys ’21 conference",
url = "https://europe.naverlabs.com/wp-content/uploads/2021/09/DuarteEtAl2021.pdf",
abstract = "Medical referrals between primary care doctors and specialists affect many aspects of patient care, such as quality of care, patient satisfaction, and health care costs. In this work, we hypothesize that the social network of medical doctors can influence the referral process. We analyse primary-specialty referrals through transaction data over medical appointments gathered between 2012 and 2017 in all hospital centers of an European private healthcare provider. In this paper we uncover patterns and hidden mechanisms in primary-specialty referrals to improve the efficiency of health care services, and propose a link prediction model for recommendation using Graph Neural Networks. First, we carried out exploratory data analysis and searched for patterns or interesting features of the data and afterwords, we created and analysed the doctor’s social network and the referrals network. We learned the representation of the referral network using a Graph Neural Network (GNN). Finally, we discussed the nodes representation of the referral network which evidences our main hypothesis. This work addresses the discovery of important patterns in medical referrals to, in future work, improve the efficiency of collaboration within organizations.",
abstractsource = "https://europe.naverlabs.com/wp-content/uploads/2021/09/DuarteEtAl2021.pdf",
abstractsourcelabel = "Published abstract source"
}