Privacy-Aware Intelligent Knowledge Management System for Performance Evaluation in the Healthcare Industry

Abstract

The healthcare sector grapples with formidable challenges in evaluating institutional performance and identifying knowledge gaps, compounded by the imperative to safeguard patient data privacy. This study introduces the Privacy-Aware Intelligent Knowledge Management System (IKMS), a novel framework that leverages machine learning to enable secure, collaborative performance assessment and systemic gap identification across healthcare institutions. The IKMS integrates privacy-preserving analytics such as differential privacy and federated learning with structured feedback mechanisms to process anonymised data exchanges and foster inter-institutional collaboration. Its core workflow encompasses the aggregation of targeted feedback from clinicians and patients, computation of a weighted Treatment Evaluation Score (TES) to quantify performance, application of Support Vector Machine (SVM) algorithms for detecting knowledge deficiencies, and delivery of tailored recommendations to support evidence-informed policy and planning. Implemented and validated in a multi-site pilot involving three National Health Service trusts, the IKMS demonstrated a statistically significant 8.5-percentage-point improvement in mean TES scores and a 14.2 percent reduction in identified gaps, all while maintaining strong privacy guarantees (ε = 0.47 differential privacy). These results highlight the system’s potential to enhance operational efficiency, strengthen evidence-based decision-making, comply with regulatory frameworks such as the General Data Protection Regulation, and ultimately promote superior patient care and institutional resilience.