Generating realistic patient data

2025-07-04

AI Summary

Patient data in healthcare are severely restricted by privacy regulations, hindering algorithm validation and result reproducibility for optimization tasks such as the Patient-to-Room Assignment (PRA) problem. Method: We propose a configurable patient instance generation framework tailored to PRA. It integrates statistical analysis of real-world data with combinatorial feasibility modeling, combining probability distribution fitting and clinical rule constraints to support customizable ward-specific characteristics (e.g., age and length-of-stay distributions), and incorporates a graphical user interface for intuitive configuration. Contribution/Results: The generated instances exhibit enhanced realism, diversity, and scenario adaptability. Experimental evaluation confirms their effectiveness in benchmarking PRA algorithms, significantly improving research reproducibility and cross-institutional comparability. This work provides a practical tool and methodological paradigm for operations research in privacy-sensitive healthcare settings.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Combinatorial Optimization
• Knowledge Representation and Reasoning: Preferences

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Developing algorithms for real-life problems that perform well in practice highly depends on the availability of realistic data for testing. Obtaining real-life data for optimization problems in health care, however, is often difficult. This is especially true for any patient related optimization problems, e.g., for patient-to-room assignment, due to data privacy policies. Furthermore, obtained real-life data usually cannot be published which prohibits reproducibility of results by other researchers. Therefore, often artificially generated instances are used. In this paper, we present combinatorial insights about the feasibility of instances for the patient-to-room assignment problem (PRA). We use these insights to develop a configurable instance generator for PRA with an easy-to-use graphical user interface. Configurability is in this case especially important as we observed in an extensive analysis of real-life data that, e.g., the probability distribution for patients' age and length of stay depends on the respective ward.

Problem

• Generating realistic patient data for testing
• Addressing data privacy in healthcare optimization
• Developing configurable patient-to-room assignment generator

Innovation

• Configurable instance generator for PRA
• Easy-to-use graphical user interface
• Combinatorial insights for feasibility analysis