Private Hospital Workflow Optimization via Secure k-Means Clustering.

Area of Science:

• Health Informatics
• Cryptography
• Organizational Workflow Optimization

Background:

• Hospital workflow optimization is complex.
• Real-time locating systems (RTLS) can track patients and staff.
• Privacy regulations restrict access to staff location data.

Purpose of the Study:

• To propose a secure solution for analyzing joined patient and staff location data.
• To enable secure clustering of staff based on patient interaction frequency.
• To address privacy concerns in hospital location data analysis.

Main Methods:

• Utilized Secure Multi-Party Computation (SMC), a cryptographic technique.
• Implemented a two-party protocol between the hospital and a trusted third party (e.g., labor union).
• Focused on securely clustering staff by analyzing frequency of patient-facing times.

Main Results:

• Developed and described a novel secure solution for analyzing sensitive location data.
• Evaluated the performance of a proof-of-concept implementation.
• Demonstrated the feasibility of secure multi-party clustering in a hospital setting.

Conclusions:

• Secure Multi-Party Computation offers a viable approach to overcome privacy barriers in hospital data analysis.
• The proposed method allows for valuable insights into staff workflow without compromising individual privacy.
• This research paves the way for privacy-preserving optimization of complex organizational workflows.