Clustering-Based Data Mining for Healthcare Decision Support Systems: A Systematic Review
Abstract
Clustering can organize heterogeneous healthcare data into clinically or operationally meaningful groups, but evidence of real decision-support use remains uneven. This systematic review examines healthcare domains, data types, clustering methods, validation practices, and implementation gaps in clustering-based Healthcare Decision Support Systems (DSS). Following PRISMA 2020, Scopus records published from 2015 to 2025 were screened using predefined eligibility criteria. Eighty-four records were identified; 36 remained after the open-access filter, 19 full-text reports were assessed, and 10 primary empirical studies were included. The studies covered fraud auditing, ICU profiling, chronic-disease risk and transition analysis, gestational-diabetes support, trauma pathways, radiology text, patient segmentation, and smart-home activity recognition. K-Means was the most frequently reported method, followed by Markov-based approaches, while other clustering techniques appeared less often. Most studies relied on retrospective, internal, or expert-based validation; only one reported prospective evaluation of a deployed web-based clinical DSS. Overall, clustering shows clear analytical value, but evidence for routine clinical deployment and patient-outcome improvement remains limited.
Related
- On the possibilities and limitations of LLM-based multi-agent systems for multi-criteria decision-making in healthcare operations management
- A bipolar fuzzy relation equation framework for clinical decision support systems
- A Hybrid Structural Modelling and Explainable Machine Learning Framework for Human–AI Strategic Decision Support Systems
- Latent Similarity Gaussian Processes: A Theory-Grounded Approach to Personalized Suicide-Risk Forecasting for Clinical Decision-Support
- CASCADE Conformal Prediction: Uncertainty-Adaptive Prediction Intervals for Two-Stage Clinical Decision Support
BibTeX
@article{lee2026clusteringbased,
title = {{Clustering-Based Data Mining for Healthcare Decision Support Systems: A Systematic Review}},
author = {Francka Sakti Lee and Catur Edi Widodo and Johanes Fernandes Andry},
journal = {Journal of Information Systems and Informatics},
year = {2026},
month = oct,
doi = {10.63158/journalisi.v8i5.1801},
url = {https://doi.org/10.63158/journalisi.v8i5.1801},
note = {CC BY},
}
RIS
TY - JOUR
TI - Clustering-Based Data Mining for Healthcare Decision Support Systems: A Systematic Review
AU - Francka Sakti Lee
AU - Catur Edi Widodo
AU - Johanes Fernandes Andry
PY - 2026
DA - 2026/10/24
JO - Journal of Information Systems and Informatics
DO - 10.63158/journalisi.v8i5.1801
UR - https://doi.org/10.63158/journalisi.v8i5.1801
AB - Clustering can organize heterogeneous healthcare data into clinically or operationally meaningful groups, but evidence of real decision-support use remains uneven. This systematic review examines healthcare domains, data types, clustering methods, validation practices, and implementation gaps in clustering-based Healthcare Decision Support Systems (DSS). Following PRISMA 2020, Scopus records published from 2015 to 2025 were screened using predefined eligibility criteria. Eighty-four records were identified; 36 remained after the open-access filter, 19 full-text reports were assessed, and 10 primary empirical studies were included. The studies covered fraud auditing, ICU profiling, chronic-disease risk and transition analysis, gestational-diabetes support, trauma pathways, radiology text, patient segmentation, and smart-home activity recognition. K-Means was the most frequently reported method, followed by Markov-based approaches, while other clustering techniques appeared less often. Most studies relied on retrospective, internal, or expert-based validation; only one reported prospective evaluation of a deployed web-based clinical DSS. Overall, clustering shows clear analytical value, but evidence for routine clinical deployment and patient-outcome improvement remains limited.
SN - 2656-5935
SN - 2656-4882
ER -