Artificial Intelligence and Bigdata

Corner Stones in Post COVID-19 Era

Yang Zhenyu, (Ph.D. candidate), Studying for doctor's degree in Kunming Medical University

Because of the hit of COVID-19, the economy of many countries in the world suffered. We must take measures to boost the economy while controlling COVID-19. However, there is a great conflict between the control of COVID-19 and the removal of restrictions on personnel mobility. How to minimise the losses caused by this conflict has become a key issue.

AI can help disease control department track and limit the spread of COVID-19

With the improvement of chip performance, trillions of calculations per second become possible. The threshold for building an ultra-high performance computing cluster is also greatly reduced. Mobile devices can all become computing nodes, and they can be used to build a huge, distributed computing network. Branch prediction calculation can be used to predict the possible movement path of each computing node by using the super high performance of this distributed computing network, to accurately predict the distribution path of COVID-19. The hypothesis results obtained through continuous real-time calculation can be verified by comparing the results obtained through epidemiological investigation, so as to train the artificial computing system. When the cumulative number of such training is enough, and the number of machine self-training is also enough, AI can accurately predict the propagation of COVID-19.

AI can help us determine the best way to manage population mobility

Population mobility has always been a key issue in the management of infectious diseases. The traditional blockade management has been proved to have huge defects in the management of COVID-19, and the same rough management method will produce huge secondary disasters as time goes on, such as psychological problems. The traditional management method will also have a huge disastrous effect on public opinion, resulting in mutual distrust of governments at all levels, distrust of the government by the people, and, worst of all, distrust among the people. This mutual distrust will cause huge trouble, and we need to make greater efforts to overcome this so-called trouble, thus falling into a vicious circle. Therefore, we can only complete the precise management of COVID-19 by ensuring population mobility as much as possible, so that it will not cause serious impact.

We can accomplish this task through the super high performance artificial intelligence network built by giving distributed computing networks as described above. We can modularise every measure that can be used to control infectious diseases. In the process of infectious disease management, the AI system is allowed to learn the consequences of this method and make self-assumptions. With the super performance of this distributed computing network, a usable AI management system can be trained soon. Match the AI prediction mentioned above with the control suggestions provided by AI, and use the AI system to predict, so that the possible results of each control measure can be displayed one by one. At the same time, the AI system can also provide management suggestions based on reality, and the precise control of COVID-19 is no longer a thorny problem.

Bigdata in epidemiological investigation

Big data technology can sort out the movement track of infected people, track the contact history of people, establish a knowledge map, and provide important information for accurate positioning of epidemic transmission path, prevention and control of epidemic spread and other aspects.

Tracking the moving track and establishing a knowledge map have become relatively mature technologies in the field of big data. In terms of location data, in addition to the travel data collected by transportation departments such as aviation, railway, highway, ferry, etc., under the premise of user authorisation, telecom operators can effectively locate the user's mobile phone location based on mobile phone signaling and other data containing geographic location and time stamp information. Internet enterprises can also call the location data of users' mobile phones through APP authorisation.

In addition, mobile travel services provided by apps such as maps, taxi taking and tourism, delivery address data in apps such as e-commerce and takeout platforms, and IP, longitude and latitude data of bank mobile payment can all be effective supplements to location data. The knowledge map can be constructed through various social platforms, communication networks, call records, transfer records and other data.

By vertically concatenating the authorised location data of mobile phone users in different time periods, the mobile trajectory can be effectively drawn. This kind of individual data can track the disease transmission path of the infected person and locate the source of infection. With the knowledge map, it can lock the people who have been contacted by the infected person, to take prevention and control measures such as isolation and treatment in time to avoid a wider spread of the epidemic.

The group data can be formed by horizontally integrating the location data of different individuals at the same time point. The use of data analysis, data mining and other technologies can accurately depict the flow direction, dynamics and scale of different categories of people who diffuse in and out across regions.

The population migration map during the epidemic can be made by using the population location data to observe the inflow and outflow of population in each city, especially the outflow direction of population in key epidemic areas. These data are helpful to locate the main regions where the epidemic situation is exported, predict the development trend of the epidemic situation in the region, predict the potential infected people in the region, and provide scientific support for disease prevention and control departments and regional governments to introduce targeted traffic control measures.

During the epidemic, the public paid close attention to the epidemic situation. How long will the epidemic spread? Will the number of infected people increase significantly? Where is the risk of infection high? When can we enter the safety period? To solve these problems, we need to find out the key influencing factors, analyse the epidemic transmission characteristics, and build the epidemic development model, in which big data can play a key role.

In addition to medical data, epidemic transmission is often affected by multi-dimensional factors such as climate, temperature, humidity, geology, transportation, social behavior and urban health. The development of big data technology enables these influencing factors to be displayed in data form, and makes multi-dimensional and large-scale data processing possible. Big data is used to realise the modeling of impact factors of tens of thousands of magnitudes, which greatly enriches the analysis dimension of epidemic development model.

The SIR model is a classic model in the infectious disease model, where S stands for Susceptible, I stands for Infected, and R stands for Removal.

The propagation process is roughly as follows: at first, all nodes are in an infectious state, and then some nodes become infected after contacting information. These infected nodes try to infect other nodes in an infectious state, or enter a removal state. In the removal state, that is, immunity, the node in the removal state will no longer participate in the propagation of information.

If mass data such as travel path flow information, social information, consumption data and exposure history are collected, and big data analysis technologies such as communication dynamics model, dynamic infection model and regression model are used, the development of epidemic situation can be more accurately predicted, and the peak inflection point of epidemic situation can be judged. In addition, it can also locate the space-time collision points according to the diagnosis order of patients and the information of close contacts, and then calculate the disease transmission path, providing a theoretical basis for the traceability analysis of infectious diseases.

Data protection and supervision

In the face of the epidemic, it is very important to ease people's anxiety. Due to the change of information access and lifestyle, search big data has become an important carrier to understand public opinion under the epidemic. The click and search behind each piece of information accurately reveal the needs and problems of the people.

The backbone of this kind of artificial intelligence calculation relying on the user's personal information is data protection and data usage supervision.

The government needs to be regulated. We need legislation to protect people's data and privacy security and ensure that the data and information obtained by the government and disease control departments can only be used for disease supervision. The data of disease management should be stored locally in the computing node and desensitised locally. The distributed computing encryption method like cryptocurrency should be adopted to protect the data and privacy security of the public and ensure that only citizens have the right to access this encrypted information.

References:
1:Zhou Y, Wang F, Tang J, Nussinov R, Cheng F. Artificial intelligence in COVID-19 drug repurposing. Lancet Digit Health. 2020;2(12):e667-e676. doi:10.1016/S2589-7500(20)30192-8
2:Alsunaidi SJ, Almuhaideb AM, Ibrahim NM, et al. Applications of Big Data Analytics to Control COVID-19 Pandemic. Sensors (Basel). 2021;21(7):2282. Published 2021 Mar 24. doi:10.3390/s21072282
3:Dong J, Wu H, Zhou D, et al. Application of Big Data and Artificial Intelligence in COVID-19 Prevention, Diagnosis, Treatment and Management Decisions in China. J Med Syst. 2021;45(9):84. Published 2021 Jul 24. doi:10.1007/s10916-021-01757-0
4:Vaishya R, Javaid M, Khan IH, Haleem A. Artificial Intelligence (AI) applications for COVID-19 pandemic. Diabetes MetabSyndr. 2020;14(4):337-339. doi:10.1016/j.dsx.2020.04.012
5:Bragazzi NL, Dai H, Damiani G, Behzadifar M, Martini M, Wu J. How Big Data and Artificial Intelligence Can Help Better Manage the COVID-19 Pandemic. Int J Environ Res Public Health. 2020;17(9):3176. Published 2020 May 2. doi:10.3390/ijerph17093176
6:Lv Y, Ma C, Li X, Wu M. Big data driven COVID-19 pandemic crisis management: potential approach for global health. Arch Med Sci. 2021;17(3):829-837. Published 2021 Mar 20. doi:10.5114/aoms/133522
7:Sadowski A, Galar Z, Walasek R, Zimon G, Engelseth P. Big data insight on global mobility during the Covid-19 pandemic lockdown. J Big Data. 2021;8(1):78. doi:10.1186/s40537-021-00474-2

Yang Zhenyu

Yang Zhenyu(Ph.D. candidate), Studying for doctor's degree in Kunming Medical University. My major research zone are critical medicine and emergency medicine. The main research direction is the application of machine learning in the medical field. I firmly believe that machine learning can bring great changes to medical technology.

Latest Issue
Get instant
access to our latest e-book
Sanofi Dubai Derma 2023