Madan, Akshay
(2025)
Wireless strategies for infectious disease control: Contact tracing and hand hygiene monitoring.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
The healthcare sector increasingly faces disruptions that demand advanced strategies, robust systems, and innovative solutions. One significant disruption has been the COVID-19 pandemic, which has claimed over 7 million lives globally. Its rapid spread and the emergence of new variants highlight the importance of preventive and screening measures such as vaccination and contact tracing (CT). CT identifies individuals exposed to infected persons, allowing timely interventions like isolation. A common method of digital contact tracing (DCT) involves using smartphones with Bluetooth to broadcast and register close contacts (phone-phone CT). However, these approaches suffer from low accuracy due to limited control over range. Also, most DCT efforts focus on direct contact, such as touching or talking, while neglecting indirect contact via contaminated surfaces or respiratory particles.
Another critical issue in healthcare is healthcare-associated infections (HAIs), which, according to the World Health Organization, are a leading cause of mortality in healthcare settings. One major contributor to HAIs is the failure of healthcare workers (doctors, nurses, etc.) to consistently adhere to hand hygiene protocols.
This factor also contributes to the transmission of infections like COVID-19 within hospitals.
Ensuring proper hand hygiene compliance (HHC) can significantly reduce the incidence of HAIs.
This dissertation addresses both of these challenges. First, it aims to enhance the accuracy of DCT while safeguarding user privacy. We use the deployment of Bluetooth-based IoT devices in public gathering spaces, such as restaurants, hospitals, and schools, to detect direct and indirect contacts. We create a simulation to study the improvements of this method over common phone-phone-based approaches and efficient strategies for placing beacons. Additionally, we extend this approach to support bidirectional tracing, identifying additional contacts arising from asymptomatic carriers. We observe that the proposed bidirectional CT outperforms existing DCT works in averting possible infections. To address the second challenge, we propose a deep learning-based system that utilizes WiFi channel state information (CSI) to monitor hand hygiene compliance. We observe that the proposed model outperforms existing time series models on an existing HHC dataset in accuracy and training time.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
7 January 2025 |
| Date Type: |
Publication |
| Defense Date: |
4 December 2024 |
| Approval Date: |
7 January 2025 |
| Submission Date: |
1 December 2024 |
| Access Restriction: |
1 year -- Restrict access to University of Pittsburgh for a period of 1 year. |
| Number of Pages: |
137 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Computing and Information > Telecommunications |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Contact Tracing, COVID-19, Bluetooth Low Energy, WiFi, Hand Hygiene Monitoring, Channel State Information, Bidirectional Contact Tracing |
| Date Deposited: |
07 Jan 2025 19:39 |
| Last Modified: |
07 Jan 2026 13:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/47149 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}