China
Africa
Explore chapters and articles related to this topic
Solutions Using Machine Learning for COVID-19
Published in Punit Gupta, Dinesh Kumar Saini, Rohit Verma, Healthcare Solutions Using Machine Learning and Informatics, 2023
Muhammad Shafi, Kashif Zia, Jabar H. Yousif
As COVID-19 is an infectious disease and it spreads due to close contacts, contact tracing is crucial for identifying potentially infected people who have been in close contact with an infected person. Most governments launched digital contact-tracing applications to identify potential infected people and break the chain of infection. Contact-tracing applications can be broadly classified into centralized and decentralized architectures. In centralized architecture, the users’ data are stored on a central server. The server uses the data for contact matching, risk analysis and sending notifications. In decentralized systems, the mobile phone itself stores the data of close contacts and performs contact matching and notification. Figure 6.2 shows typical centralized and decentralized contact-tracing architectures.
Machine Learning Implementations in COVID-19
Published in Chhabi Rani Panigrahi, Bibudhendu Pati, Mamata Rath, Rajkumar Buyya, Computational Modeling and Data Analysis in COVID-19 Research, 2021
Kabita Kumari, S.K. Pahuja, Sanjeev Kumar
The most crucial step for controlling coronavirus’s spread is contact tracing, as the virus spreads from one person to another through saliva and droplets (www.who.int). Contact tracing plays a vital role in the healthcare system. It will help identify and manage the people with COVID-19 and can suppress the outbreak of a pandemic throughout a population (Lalmuanawma, Hussain, and Chhakchhuak 2020). Nowadays, digital contact tracing methods are used by many countries, notably South Korea and Singapore (Wong, Leo, and Tan 2020), such as mobile data tracing, GPS (Global Positioning System), proximity tool Bluetooth, etc. These methods allow quicker processing of data than non-digital systems. The digital tracing process employs machine learning and artificial tools for the analysis of the disease. Several countries have employed ML and AI in digital tracing for infectious chronic wasting disease (Rorres et al. 2018) using centralized, decentralized, or hybrid techniques to minimize traditional, labor-intensive, and manual tracing methods. One study (Ferretti et al. 2020) highlighted the challenges and voluntariness of COVID-19 tracking apps (CTAs) that provide information about testing or advice for self-isolation from healthcare experts. A schematic diagram for COVID-19 contact tracing based on apps is shown in Figure 1.6.
Infection Tracing in i-Hospital
Published in Ricardo Armentano, Robin Singh Bhadoria, Parag Chatterjee, Ganesh Chandra Deka, The Internet of Things, 2017
Mimonah Al Qathrady, Ahmed Helmy, Khalid Lmuzaini
Contact tracing is an important mean of controlling infectious diseases. Armbruster and Brandeau (2007a) developed a simulation model for contact tracing and used it to explore the effectiveness of different contact-tracing policies in a budget-constrained setting. A simulation model of contact tracing is used to evaluate the cost and effectiveness of different levels of contact tracing (Armbruster & Brandeau, 2007b).
Campaigning for the greater good? – How persuasive messages affect the evaluation of contact tracing apps
Published in Journal of Decision Systems, 2022
Covid-19 contact tracing apps seek to facilitate the early detection and interruption of infection chains by tracking personal encounters with other individuals and by providing early warnings after contact with an infected person. Most tracing apps use Bluetooth technology since it preserves more privacy in comparison to GPS (Ciucci & Gouardères, 2020; O’Neill et al., 2020). Some of the tracing apps also rely on additional interface technology by Apple and Google. When tracing apps are active, smartphones search for other smartphones on which the app is installed. When two individuals approach each other with the same app, automatically developed identifiers are exchanged and stored either centrally or locally. There are some country-specific variations, but many contact tracing apps set the threshold to a distance of less than two metres and a time of more than 15 minutes during which persons would be exposed to a significant risk of infection. Only encounters that meet both the time and distance conditions are traced. Disclosing a known infection in the tracing app remains the individual’s responsibility (Ciucci & Gouardères, 2020).
Privacy risk in contact tracing systems
Published in Behaviour & Information Technology, 2023
In absence of a pharmaceutical treatment, contact tracing is a core public health strategy used to control infectious disease transmission. The early global outbreak of the novel coronavirus COVID-19 spurred automated contact tracing methods as a means to leverage technological advances. Contact tracing serves two distinct yet interrelated public health objectives: (1) identifying, notifying, and treating infected individuals; and (2) mapping the epidemiological transmission and route of disease spread. The present study set out to answer two research questions. First, what privacy risks are inherent in contact tracing in general? Second, how do privacy risks differ between conventional manual and modern automated contact tracing systems? We address each question next.
Evaluating the Performance of Wearable Devices for Contact Tracing in Care Home Environments
Published in Journal of Occupational and Environmental Hygiene, 2023
Kishwer Abdul Khaliq, Catherine Noakes, Andrew H. Kemp, Carl Thompson
Contact tracing is a proven method of mitigating the spread of infectious diseases, and digital technologies could help to ease this process. This study evaluated the performance of a wearable BLE based solution combined with LoRaWAN technology, using experiments in a range of different scenarios to quantify the reliability and accuracy of the system. The results show that the system performance depends on the environment, with the following key conclusions drawn:The overall average system’s contact detection success rate was measured as 75.5%. When wearables were used as per the guidelines the rate increased to 81.5%., but when obstructed the rate was only 64.2%.The difference in accuracy between different forms of the BLE devices (watch, fob or card) was small, suggesting that all types of devices can be used in a similar way.Everyday obstructions such as placing the device in a bag or under a scarf significantly reduced the accuracy of the system, with calculated distances greater than physical distances. This is likely to lead to false negative contacts when used with a particular time-distance contact threshold.Contacts measured outdoors were accurate in terms of distance but had a greater variance and a lower contact detection success rate than for experiments conducted indoors.Bluetooth propagation through walls and doors means that false positive contacts can be recorded by the system even when devices are located in different rooms.