Data-driven test strategy for COVID-19 using machine learning: A study in Lahore, Pakistan

doi:10.1016/j.seps.2021.101091

Socio-Economic Planning Sciences

Volume 80, March 2022, 101091

https://doi.org/10.1016/j.seps.2021.101091 Get rights and content

Highlights

•: Findings revealed the failure of a current test strategy.
•: A data-driven test strategy for COVID-19 based on machine learning was proposed for policy-making insights.
•: The proposed strategy was much more efficient under strictly limited test capacity.
•: Long-term data collection was not prerequisite for the conduction of the data-driven strategy.
•: Even for different subareas of a city, the strategy driven by local data was likely to be optimal.

Abstract

Aims

We aimed at giving a preliminary analysis of the weakness of a current test strategy, and proposing a data-driven strategy that was self-adaptive to the dynamic change of pandemic. The effect of driven-data selection over time and space was also within the deep concern.

Methods

A mathematical definition of the test strategy were given. With the real COVID-19 test data from March to July collected in Lahore, a significance analysis of the possible features was conducted. A machine learning method based on logistic regression and priority ranking were proposed for the data-driven test strategy. With performance assessed by the area under the receiver operating characteristic curve (AUC), time series analysis and spatial cross-test were conducted.

Results

The transition of risk factors accounted for the failure of the current test strategy. The proposed data-driven strategy could enhance the positive detection rate from 2.54% to 28.18%, and the recall rate from 8.05% to 89.35% under strictly limited test capacity. Much more optimal utilization of test resources could be realized where 89.35% of total positive cases could be detected with merely 48.17% of the original test amount. The strategy showed self-adaptability with the development of pandemic, while the strategy driven by local data was proved to be optimal.

Conclusions

We recommended a generalization of such a data-driven test strategy for a better response to the global developing pandemic. Besides, the construction of the COVID-19 data system should be more refined on space for local applications.

Keywords

COVID-19

Test strategy

Policy making

Machine learning

Logistic regression

Time series analysis

Spatial analysis

Chuanli Huang was born in Wuhan, China, in 1993. He received his Bachelor's Degree at State Key Laboratory of Fire Safety Science of University of Science & Technology of China (2016), and has been doing PhD in University of Science & Technology of China, and City University of Hong Kong after graduation. His research area includes urban emergency response, resource allocation optimizing, and evacuation dynamics. He has participated in several national or provincial research projects including the National key Research and development program.

Min Wang was born in Tongcheng city, Anhui Province, China in 1993. She is a physician received the Medical Bachelor's degree from Jining Medical College in 2017 and clinical degree of medical Master from Soochow University in 2020. From 2012 to 2020, she has been working in the Third-grade A hospital. At present, she has mastered the diagnosis and treatment of common diseases of different clinical departments. Besides, She has published 2 scientific and technological core journals as the first author and has participated a university-level research project. Her research interests include standardized treatment of malignant tumors, nuclear emergency injury treatment and health protection against occupational diseases.

Warda Rafaqat born in Pakistan did her sixteen years education in geographical information systems from Punjab University College of Information Technology, Lahore Pakistan. She worked with Urban Unit and Data Solutions Private Limited for three years. Currently doing her Masters from USTC Hefei. GIS, RS, ML are her interests and implementation of the fields for Safety and Management.

Salman Shabbir born in Pakistan did his sixteen years education in geographical information systems from Punjab University College of Information Technology, Lahore Pakistan. He worked with Urban Unit and Punjab Disaster Management Authority. Currently working with PUnjab Information Technology Boards. He worked in Punjab Dengue Program and worked under different projects for Punjab Government where he used GIS techniques.

Liping Lian is a postdoctoral fellow in Peking University Shenzhen Graduate School. She received her PhD at University of Science & Technology of China (2018) and City University of Hong Kong. She is interested in pedestrian evacuation and dynamic and publish six scientific papers in this field. She is responsible for one project from National Natural Science Foundation of China and one project from China Postdoctoral Science Foundation.

Prof. Jun Zhang is the professor of University of Science & Technology of China. He received his PhD at University of Wuppertal, Germany(2012), and was a Postdoctoral Fellow of Research Centre Juelich, Germany (2012–2017.)His research area includes pedestrian traffic dynamics, pedestrian and evacuation dynamics, capacity assessment on emergency treatments, and application of virtual reality technology. He is scientific committee member of the International Conference on Pedestrian and Evacuation Dynamics (2018-), editorial board member of Collective Dynamics (2019-), member of International Association of Fire Safety Science(IAFSS), and communication expert of National Natural Science Foundation of China. His work have been supported by two projects from the National Natural Science Foundation of China, a project from National Key Research and Development Program of China (Grant No. 2018YFC0808600), a project from the Anhui Provincial Natural Science Foundation and a project from the Fundamental Research Funds for the Central Universities. Dr. Zhang has published more than 40 scientific papers

Prof. SM Lo is a professor of the Department of Architecture and Civil Engineering, City University of Hong Kong. He received his PhD in Architecture from the University of Hong Kong. His main research interests include spatial planning for pedestrian movement, decision support system, fire protection engineering and evacuation, urban planning and construction management. He has been the Principal Investigator for many research grants including 14 Competitive Earmarked Research/ GRF Grants from the Research Grant Council of HKSAR for studying evacuation modeling, pedestrian flow modeling, fire risk analysis, decision support system, wayfinding modeling, intelligent understanding of CAD plans, human behavior in fire, effect of fire legislation on the society, large-scale evacuation modeling and disaster prevention, rail tunnel evacuation and etc. He is also a Co-PI for a Theme-based Research Project: Safety, Reliability, and Disruption Management of High Speed Rail and Metro Systems. He has published more than 150 articles in SCI/ SSCI listed journals. He is also an Authorized Person registered under Hong Kong Buildings Ordinance.

Prof. Weiguo Song is the professor of University of Science & Technology of China. He received his PhD also at University of Science & Technology of China (2001). His has been focusing on pedestrian and evacuation dynamics, fire monitoring with remote sensing, self-organized criticality of fire system as well as performance-based fire protection design of buildings. Prof. Song has presided over three projects supported by the National Natural Science Foundation of China, three projects supported by the National Key Technologies R & D Program of China, one project supported by the Ministry of Science and Technology of China, and one project supported by the Ministry of Education of China. Moreover, Prof. Song has participated in many research projects supported by the National Basic Research Program of China (also known as the “973” Program), and the National Key Technologies R & D Program of China in his capacity as a key researcher. Prof. Song entered the list of recipients supported by the “New Century Talents Supporting Program” of the Ministry of Education in 2008. Prof. Song has authored over 60 research papers, published one book, and owned copyrights of four computer softwares. Prof. Song is a member of International Association of Fire Safety Science(IAFSS), member of China Association of Public Safety Science and Technology (2012-). He is also a reviewer of Risk Analysis, Physica A, Building and Environment, Ecological Modelling, IEEE Transactions on Intelligent Transportation Systems, International Journal of Remote Sensing etc.

¹: co-first authors.

View Abstract