key: cord-1009266-9wd1jrsi
authors: Jadhav, Jagannath; Rao Surampudi, Srinivasa; Alagirisamy, Mukil
title: Convolution Neural Network Based Infection Transmission Analysis on Covid -19 Using GIS and Covid Data Materials
date: 2021-03-04
journal: Mater Today Proc
DOI: 10.1016/j.matpr.2021.02.577
sha: e20f4b58ab25b04f345ae676b39073d740fcbcfa
doc_id: 1009266
cord_uid: 9wd1jrsi

Towards the improvement of predicting and analyzing the infection transmission, a novel CNN (Convolution Neural Network) based Covid Infection Transmission Analysis (CNN-CITA) is presented in this article. The method works based on both GIS data set and the Covid data set. The method reads all the data from the data sets. From the remote sensing data, the method extracts different climate conditions like temperature, humidity, and rainfall. Similarly from Global Information System data set, the locations of the peoples are fetched and merged. The merged data has been split into number of time frame, at each condition, the data sets are merged. Such merged data has been trained with deep learning networks which support the search of person location and mobility. Based on the result and the data set maintained by the governments, the infection transmission rate has been measured on region basis. In each region of movement performed by any person, the method computes the infection Transmission Rate (ITR) in two time window as before and after. According to the infection rate and ITR value of different region, a subset of sources are selected as vulnerable sources. The method produces higher performance in predicting the vulnerable sources and supports the reduction of infection rate. Index Terms: CNN, CNN-CITA, Regional Transmission, Infection Rate, ITA, ITS, GIS, Remote Sensing Data.

In the recent times, the novel Corona virus has been identified as the most threatening disease which has been affected by huge world population. The virus has been mentioned as spread from china as it has been first identified in Wugan a city in China. As it spreads from the fish market in that city, and continuous to spread to many countries. The human society facing huge challenge from novel corona virus which is being spread throughout the world more than 230 countries. Still there is no solution has been identified for the problem of Covid-19. As the virus is pandemic which is capable of getting transmitted to person to person in higher rate, containing the rate of infection has become a huge challenge.

According to the media information, the most world countries has enforced lock down to block their people mobility due to the reason that the virus spreads between human to human either by contact or through some water particles expose through the mouth of the human. Due to the infection, there mortality rate has been hiked in Europe countries like Germany, Brazil and many more. The USA (united states of America) has been identified as the most hit country by the Covid-19 virus. To secure the people from the virus various activities has been taken by different countries but still its rate of infection is getting increased every day. The government of any country is maintaining the details of infected peoples and their location with their mobility.

By pinpointing the remote sense data with GIS data set at current time frame, their mobility of different infected peoples who supports the spread of pandemic can be analyzed. According to the data available in both remote sensing and GIS, both can be merged to analyze the transition of infection to different geographic region. This would support the detection of serious carriers and would help to block such serious carriers. Because not all the infected human is capable of transmitting to all others but by identifying the serious carriers, the rate of pandemic can be reduced.

Towards the scope, a CNN based infection transmission analysis model is presented. The method uses both geologic and pandemic data in the transmission analysis. The model split the geographic area into number of regions and measure infection rate on various areas at different time stamp. According to the GIS data and mobility of users, the method computes the infection transmission support (ITS) to find the vulnerable users. Similarly, the CNN is adapted for various problems of prediction and the same can be used towards transmission analysis. The data belongs to GIS can be trained and according to the presence of particular person in particular region, the infection induction rate (IIR) can be measured to compute the ITS. To support this, the remote sensing data has been used with the GIS data. The detailed approach is presented in detail in the next sections.

There are a variety of methods explored in the literature to classify susceptible sources for different diseases. This section outlines the collection of strategies associated with the problem.

Remote Sensing in Human Health: A 10-Year Bibliometric Analysis[1] offers an overview of the state of science on the use of remote sensing evidence in human health and highlights the most involved partners, e.g. writers and organisations in the area, to advise future new collaborative research groups. The goal of this research is to analyse, track, map, schedule, distribute and locate government PHCCs in the selected study area in order to incorporate remote sensing and GIS to manage primary health care centers [2] . The paper also focuses on the follow-up of supply and demand in PHCCs; maintaining various health care (HC) facilities in compliance with the requirements and expectations of the Ministry of Health and Population (MOHP) for the incorporation of remote sensing data and geographic information system GIS from a regional and HC viewpoint for the best treatment facility.

Geographical Information Systems (GIS) should be used as a means of preparing for health care providers to reduce the backlog of patients waiting to be seen. In order to determine how useful the associations between Kano Metropolis neurons can be, 6 central blue E-type neurons and 6 central yellow N-type neurons have been picked. The report was performed with ArcView GIS 3.2a for the ArcView GIS analysis and the findings of the research reveal that most health facilities are concentrated between Kano Municipal and Nassarawa and Tarauni. Remote-Sensing Systems for Environmental Health Research [4] , address that the International Space Station would acquire three new devices, one that will monitor how winds work across the planet, one that will quantify clouds and aerosols (particles floating throughout the atmosphere)-two factors that remain challenging to forecast in climate change models-and one that will take worldwide, long-term action. With respect to temperatures in a wide range of climate zones across Africa, the best data shows that short-term temperature variations can have a direct effect on pathogen transmission.

Climate Information for Public Health: the role of the IRI Climate Data Library in an Interconnected Knowledge System [7] , discuss that the Climate Information for Public Health Initiative of the International Research Institute for Climate and Society (IRI) aims to improve the capacity of the public health sector to recognise, use and demand relevant climate data and climate information. The use of remote sensing as a monitoring instrument for endemic diseases in Brazil [8] poses the characteristics and potentialities of remote sensing as a valuable environmental observation method for applied endemic management research in Brazil. [8] Onboard satellite sensors allow a comprehensive study of a given territory, delivering in-depth spatial and temporal information on an extremely broad scale and at a regional level. When [9] the point estimation of SARS-CoV-2 seroprevalence in asymptomatic individuals over time was reviewed [10] , two cohort studies were reviewed.

The author discusses two separate aspects in which the hygiene effect could have an influence on our bodies. In the one hand, this will allow our body to remain safe and even develop, but on the other hand, it may also create other issues. Here, it is proposed that the disruption of microbial sharing associated with dysbiosis (loss of bacterial diversity associated with a microbiota imbalance with deleterious consequences for the host) can worsen the prognosis of COVID-19 disease. In Article [11] , the author addresses that the virus has spread around the globe and In the previous article, [12] the author explains and describes in detail the different clinical dimensions of the Covid pandemic that are spread by inhalation or interaction with infectious droplets and, for example, the incubation time is between 2 and 14 days. Diseases that also contain these symptoms are generally fever, cough, sore throat, shortness of breath, weakness, malaise, among others. The most prevalent cause of disease is cold and mild influenza, where patients suffer nausea, vomiting and diarrhoea, fever, headache, cough, sore throat, exhaustion, weak appetite, ambiguous muscle and joint aches, and sore throat. Any of the symptoms can grow into pneumonia, acute respiratory distress syndrome (ARDS) and multi-organ failure in people with many diseases. The reviewer of the report [13] addresses that COVIID-19 is an often occurring virus that could be responsible for pandemics. We can see that, since the majority of cases identified for this disease require more serious treatments, the mortality rate for this disease is poor. In [14], the author addresses AI and big data, then describes AI as a method for detecting COVID-19, then highlights numerous problems and problems, such as the lack of a consistent reference point and the lack of an in-depth understanding of COVID-19, then concludes with a set of communications guidelines aimed at effectively reducing the COVID-19 problem. In [15] , the author introduces a new interactive visual interface that displays and compares the pace of spread of the COVID-19 pandemic over time across various countries. We surveyed current visualisation methods used in different websites and media outlets and adopted the use of a knee detection algorithm that divides exponential distribution across several linear components. It was found that during the pandemic, people in European countries travelled five times to reach their favourite destinations, that they wanted to fly to their destination five times, that they selected their destination five times (5) , and that they played a random-combination of their U-VNO-VNO journey within their origin-transportation-destination (OTD) network.

In [17] , the author used details on the outbreak curve of cases of 2019-nCoV from January 10 to January 24 and used the exponential growth model to predict the number of cases of 2019-nCoV in mainland China from January 10 to January 24, 2020. Using the evolution rate μ and the number of serial intervals (such as 400 SI), we calculated the measurement unknown as μ, the evolution rate, using a number such as 400 SI of two other well-known coronaviruses. In his article in the "Scandinavian Journal of Infectious Diseases" the author adapted the recorded serial interval (mean and standard deviation) to the gamma distribution and applied the "earlyR" package in R to approximate R0 at the early stage of the outbreak of COVID-19. The "projections" kit in R was used to predict the predicted projected disease trajectories and potential daily occurrence by fitting the current daily incidence data, the distribution of serial intervals, and the projected R0 into a model based on the assumption that daily investigators roughly follow Poisson's daily infectious distribution.

In [19] , suggest a computational model for the epidemic of Wuhan COVID-19 that takes into account human (e.g. behavioural responses such as accommodation) and government activities such as restricted holidays and quarantine for infected people. In this publication (under the title "Tick bite poisoning: a fascinating new world of toxicity"), the scientists collected the dates of onset of the disease in cases of tick bites that could lead to tick-borne diseases. All the proof they examined was classified on a scale of 1 to 7 by the review committee in order to convert it into a grade. These ratings were then extended to a subset of pairs with the highest reporting certainty.

We correct for the left truncation of the data by using an approximation that is different from the values we have previously measured. [21] records a newly discharged asymptomatic patient with COVID-19 who has been screened positive for SARS-CoV-2. The question is that COVID-19 was the cause of the latest hospitalisation of the patient and other asymptomatic hard-to-diagnose patients. The explanation that 987 new coronavirus cases were identified by Iran as of 1 March 2020 is that, as of 1 March, the country had reported 54 related deaths. We studied the epidemiology of COVID-19 in several Middle East countries and found that at least six neighbouring countries registered imported cases. In this report, data on air transport and the number of cases smuggled from Iran to other Middle Eastern countries were used to approximate the number of COVID-19 cases in Iran. The overall number of cases in Iran was projected to be 16,533 (95% confidence interval: 5925-35,538) by 25 February 2020 before the UAE and other Gulf Cooperation Council countries suspended inbound and outbound flights from Iran.

In [23] , the author addresses that the number of new cases of coronavirus (COVID-19) tends to rise worldwide, and the discrepancy between data from China and statistical estimates of occurrence based on cases diagnosed outside of China suggests that a large number of cases remain under-diagnosed (Nishiura et al., 2020a) . In [24] , the author makes it clear that the viral load dynamics between imported and non-imported patients with clinical characteristics of COVID-19 are distinct. While it will start mostly as pneumonia, there is a small risk of SARS- Table 4 , which shows the proposed CNNCITA approach has produced less time complexity than other techniques. 

GIS for health services

Remote-Sensing Applications for Environmental Health Research

Modeling the effects of weather and climate change on malaria transmission. Environ Health Perspect 1185620-626

Implications of temperature variation for malaria parasite development across Africa

Climate information for public health: the role of the IRI climate data library in an integrated knowledge system

Remote sensing as a tool to survey endemic diseases in Brazil, SCIELO

SARS-CoV-2 Seroprevalence among a Southern U.S. Population Indicates Limited Asymptomatic Spread under Physical Distancing Measures

The Social Distancing Imposed To Contain COVID-19 Can Affect Our Microbiome: a Double-Edged Sword in Human Health, American society of microbiology

An Early Pandemic Analysis of SARS-CoV-2 Population Structure and Dynamics in Arizona, American society of microbiology

A Review of Coronavirus Disease-2019 (COVID-19)

The Indian perspective of COVID-19 outbreak

Artificial Intelligence (AI) and Big Data for Coronavirus (COVID-19) Pandemic: A Survey on the State-of-the-Arts

COVID-19 Data Visualization through Automatic Phase Detection

Understanding the Impact of the COVID-19 Pandemic on Transportationrelated Behaviors with Human Mobility Data

Preliminary estimation of the basic reproduction number of novel coronavirus (2019-nCoV) in China, from 2019 to 2020: A data-driven analysis in the early phase of the outbreak

Estimation of the reproductive number of novel coronavirus (COVID-19) and the probable outbreak size on the Diamond Princess cruise ship: A data-driven analysis

A conceptual model for the coronavirus disease 2019 (COVID-19) outbreak in

China with individual reaction and governmental action

Serial interval of novel coronavirus (COVID-19) infections

SARS-CoV-2 turned positive in a discharged patient with COVID-19 arouses concern regarding the present standards for discharge

Preliminary estimation of the novel coronavirus disease (COVID-19) cases in Iran: A modelling analysis based on overseas cases and air travel data

Estimation of the asymptomatic ratio of novel coronavirus infections (COVID-19)

Clinical features and dynamics of viral load in imported and non-imported patients with COVID-19