HealthPass: a contactless check-in and adaptive access control system for lowering cluster infection risk in public health crisis

G Luo, Yufei Wang, Liang Hong, Xin He, Jiaru Wang, Shen Qu, Cheng Wang, Longbiao ChenFrontiers in Public Health 2024 · DOI

Abstract

The system has demonstrated its capacity to reduce infection risks by adapting visitor entry procedures based on individual risk factors and venue conditions. Our results suggest that the integration of a dynamic queuing model and real-time data analysis can effectively manage the flow of visitors while ensuring public health safety.

Introduction

During public health crises like the COVID-19 pandemic, authorities face the critical challenge of balancing infection control with the necessity of keeping essential venues—such as schools, hospitals, and grocery stores—operational. Traditional access control methods, including paper logs and basic health code applications, often fail to account for real-time venue capacity or risk surface transmission. HealthPass addresses this by providing a dynamic access management solution that regulates entry based on individual infection risk and real-time venue crowdedness, all while preserving user privacy.

Methodology

HealthPass replaces binary "allow/deny" access rules with a dynamic, priority-based queuing system. The workflow involves several integrated components:

  1. Contactless Check-in & Self-Reporting: Visitors scan a QR code and complete an anonymous self-declaration form covering residence, mobility, and health symptoms.
  2. Visitor Liability Profiling: Data is mapped onto a 6-axis radar chart to calculate a "liability score" (risk level). The geometric area of this chart determines the visitor's risk level.
  3. Venue Availability Monitoring: Safety is evaluated using the SEIR (Susceptible-Exposed-Infectious-Recovered) model, considering high-risk visitors currently inside, propagation risk from neighboring venues, and overall crowdedness.
  4. Dynamic Access Control: Utilizing an M/M/s queuing model with non-preemptive priority, the system assigns wait times. A visitor's liability score dictates their priority, while the venue's real-time availability dictates service capacity.
System overview showing the check-in process, risk profiling, and venue monitoring.

Data and Simulation

The system was validated using a Campus Deployment Dataset consisting of 2,044,719 check-in instances from 29,791 anonymized users across 52 university venues. Additionally, a 3D simulation platform modeled viral spread using laser-scanned campus layouts to test the system's efficacy in various scenarios.

Key Results

Simulation results indicate that HealthPass significantly "flattens the curve." With a 5% initial infection rate, the system reduced the peak number of active cases by 63.82% (from 1,302 down to 471) and delayed the infection peak from day 11 to day 19 compared to no intervention.

Real-time visitor monitoring for different venues over a 24-hour period.

Furthermore, the system achieved a System Usability Scale (SUS) score of 84.59, reflecting high user acceptance. During real-world deployment, HealthPass successfully managed crowd surges during graduation events by dynamically extending wait times to maintain safe occupancy.

Discussion and Conclusion

While HealthPass relies on the honesty of self-reported data, it introduces a robust framework for managing public health risks through dynamic queuing rather than hard lockouts. The use of radar maps for risk scoring and graph databases for networked venue risk allows for a nuanced approach to safety. Future work will explore expanding the system to non-pandemic applications, such as hospital scheduling and large-scale event crowd control, while integrating deeper behavioral analysis to detect abnormal patterns.

Summary updated

Abstract

The system has demonstrated its capacity to reduce infection risks by adapting visitor entry procedures based on individual risk factors and venue conditions. Our results suggest that the integration of a dynamic queuing model and real-time data analysis can effectively manage the flow of visitors while ensuring public health safety.
BibTeX
@inproceedings{luo2024healthpass,
  author = {G Luo and Yufei Wang and Liang Hong and Xin He and Jiaru Wang and Shen Qu and Cheng Wang and Longbiao Chen},
  title = {{HealthPass: a contactless check-in and adaptive access control system for lowering cluster infection risk in public health crisis}},
  booktitle = {Frontiers in Public Health},
  year = {2024},
  doi = {10.3389/fpubh.2024.1448901},
}
Figure