Coupling Neural Networks Between Clusters for Better Personalized Care

Simple item page
dc.contributor.authorKraus, Mathias
dc.contributor.authorHambauer, Nico
dc.contributor.authorMüller, Kristina
dc.contributor.authorKröckel, Pavlina
dc.contributor.authorUlapane, Nalika
dc.contributor.authorDe Caigny, Arno
dc.contributor.authorDe Bock, Koen
dc.contributor.authorCoussement, Kristof
dc.date.accessioned2023-12-26T18:41:47Z
dc.date.available2023-12-26T18:41:47Z
dc.date.issued2024-01-03
dc.identifier.doi10.24251/HICSS.2023.438
dc.identifier.isbn978-0-9981331-7-1
dc.identifier.other0d0ed4a8-4af3-47bc-b684-d5fb2033351b
dc.identifier.urihttps://hdl.handle.net/10125/106821
dc.language.isoeng
dc.relation.ispartofProceedings of the 57th Hawaii International Conference on System Sciences
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectPersonal Health Management with Digital Solutions
dc.subjectmachine learning
dc.subjectneural networks
dc.subjectoptimization
dc.subjectpersonalized care
dc.titleCoupling Neural Networks Between Clusters for Better Personalized Care
dc.typeConference Paper
dc.type.dcmiText
dcterms.abstractPersonalized healthcare powered by machine learning (ML) is at the forefront of modern medicine, promising to optimize treatment outcomes, reduce adverse effects, and improve patient satisfaction. However, simple ML models generally lack the complexity to accurately model individual characteristics, while powerful ML models require large amounts of data, which are often unavailable in the healthcare domain. We address this problem with cluster-level personalization. In this method, similar patients are grouped into clusters and a local ML model is trained for each cluster. Since the amount of patient data to train ML models naturally decreases for each cluster, we introduce a novel objective function called "coupling" that allows information to be shared between clusters, so that smaller clusters can also benefit from information from larger clusters, thereby improving patient outcome prediction. Our method provides a compromise between a single global model for all patients and completely independent local cluster models. We show that coupling leads to statistically significant improvements on a simulated and a real-world dataset in the context of diabetes.
dcterms.extent10 pages
prism.startingpage3627

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
0356.pdf
Size:
988.1 KB
Format:
Adobe Portable Document Format
Download

Collections

Personal Health Management with Digital Solutions