key: cord-0796701-gl54n1xm authors: Pang, Sulin; Wu, Jiaqi; Lu, Yinhua title: Thermodynamic imaging calculation model on COVID-19 transmission and epidemic cities risk level assessment—data from Hubei in China date: 2021-01-09 journal: Pers Ubiquitous Comput DOI: 10.1007/s00779-020-01478-0 sha: f056a268f5d545798eb34a024eebeca067288ca4 doc_id: 796701 cord_uid: gl54n1xm Novel coronavirus pneumonia in 17 city (Hubei) provinces was analyzed by using the principle of thermodynamics. A thermodynamic imaging model of infectious diseases was established to calculate the cumulative superimposed density of epidemic in 17 cities (prefectures). An evaluation rule of urban risk grade is established and evaluates the COVID-19 risk of 17 cities. The results show that (1) the higher the superimposed density of urban epidemic, the more infected people. (2) In the incubation stage, the thermodynamic imaging shows a point distribution, random walk, and outward diffusion trend. In the initial stage, the color of thermodynamic imaging gradually deepened and the range gradually expanded. During the burst stage, the thermodynamic imaging color deepens rapidly and the scope expands rapidly. In the stable stage, the thermodynamic imaging color becomes darkest and the range is extended to the pole. (3) According to the situation of COVID-19 transmission in Hubei Province, the cumulative superimposed density of Wuhan epidemic is far more than 10,000, ranking as “highest-risk.” Xiaogan and other 10 cities have a cumulative superimposed density within the range of [1000, 10,000], ranking as “high-risk.” Shiyan and other 5 cities have accumulated superimposed density values within the range of [100, 1000], ranking as “medium-risk.” Shennongjia cumulative superimposed density value is less than 100, and the level is “low-risk.” A novel coronavirus pneumonia was found out in Wuhan, China's Hubei Province at the end of 2019, and gradually transmitted to most countries in the world. The novel coronavirus pneumonia is a novel coronavirus infection named COVID-19 caused by acute respiratory infectious diseases. At 24:00 on 1 May 2020, the Chinese mainland has reported 82,875 confirmed cases, and 4633 cases (http://www.nhc.gov. cn/xcs/yqtb/202005/3e2bc587447f458c809f10897bb4327e. shtml) have reported deaths [1] . Outside of China, the global situation is getting worse. As of 10:00 CET (Central European Time) on 1 May 2020, the total number of confirmed cases in Spain was 213,435, Italy 205,463, the USA 1,035,353, and the global total number of reported deaths was 224,172. The cumulative number of confirmed cases was 3,175,207 (https:// www.who.int/docs/default-source/coronaviruse/situationreports/20200501-covid-19-sitrep.pdf?sfvrsn=742f4a18_2) [2] . Since the outbreak began, the global public health service system has been severely impacted. The clinical characteristics of COVID-19 patients' infection are mainly based on the basic data such as the epidemiological characteristics, laboratory examination data, and clinical characteristics of 41 patients who reported the infection [3] . Human-to-human transmission has occurred among close contacts since mid-December 2019 [4] . Unlike MERS-COV and SARS-COV, 2019-nCoV is the seventh member of the coronavirus family that infects humans [5] , and COVID-19 grows twice as fast as MERS and SARS [6] . The COVID-19 is a group-onset disease, which is more likely to cause complications in elderly males and, in severe cases, acute respiratory distress syndrome [7] . Most of the patients developed symptoms such as fever, cough, myalgia, and obstruction of the upper respiratory tract before admission [8] . In general, since the outbreak of novel coronavirus, researchers have analyzed the medical characteristics, treatment methods, prevention, and control management of novel coronavirus pneumonia from different disciplinary perspectives and different research methods, and some scholars have analyzed the mechanism and law of medical transmission of COVID-19 disease from individual to individual based on the medical characteristics, but few scholars as a whole combined with geographical features, population density, population distribution analysis of the image characteristics of COVID-19 transmission, geography characteristics, such as less scholars from the perspective of thermodynamics principle of image analysis, the transmission of the COVID- 19 . The field of thermodynamics, a method used to record and identify the interaction risks between production safety personnel and heavy construction equipment, analyzes the spatial and temporal GPS data, and displays the disaster and risk index in the form of thermal map, which can prevent and reduce the occurrence of construction accidents to some extent [9] . A new heat map method combines the traditional heat map method with the fire risk theory to establish a linear model of fire risk, which provides a reference for evaluating the relationship between regional fire risk and population density [10] . In daily life, thermal maps are also used to evaluate the travel characteristics of public transport, intuitively reflect the spatial and temporal characteristics and patterns of the emerging demands of public transport, and provide a set of practical visual analysis methods for the determination of the travel characteristics of public transport [11] . The use of Baidu thermal map data to analyze the traffic and usage of the park is more accurate than the Weibo check-in data to estimate the usage of the park [12] . In the field of intelligent navigation research, a new robot path discrimination algorithm based on thermal map uses a fixed external camera to identify the congestion to generate a noncongestion path, thus improving the navigation performance of the robot and providing a safe non-blocking path for robot navigation [13] . In the diagnosis of breast cancer risk in the study, model of machine learning method is applied in the diagnosis of breast tumor [14] ; this model is mainly based on machine learning method to diagnose cancer risk, but this article is mainly based on the thermodynamic imaging model theory and principles to COVID-19 risk grading model, different from the traditional machine learning method, is advantageous to the visual observation of COVID-19 situation changes, and has certain innovation. In other studies of common diseases, there are also research teams that build earlywarning risk models based on rough set theory [15] , so as to improve the early treatment and diagnosis of cardiovascular diseases. Similarly, this paper mainly constructed the model based on the thermodynamic imaging principle, and calculated the COVID-19 risk level of the region based on the most original data. Different from the rough set theory, the thermodynamic image principle has higher visualization value and imaging value. A smart thermometer and mobile application was proposed for influenza prediction, but we mainly proposed a thermodynamic imaging to monitor and predict epidemiological trends, resulting in a more intuitive image [16] . An infectious disease analysis framework was proposed for analyzing the dynamics of infectious diseases and proposing interventions [17] . But this paper mainly uses the thermodynamic imaging calculation model to analyze the epidemic dynamics, which is different from the framework they proposed for the analysis of infectious diseases. Overall, thermodynamic imaging technology and ideas are widely used in intuitive visual image of the industries and fields, but few scholars apply heat map to thought to major public health emergencies such as the transmission of infectious disease model study, major sudden outbreak of infectious diseases mainly through person-to-person transmission, through thermal image intuitive display can better restore its regional transmission mechanism and the law, to a major public health emergency cut off provides a new thinking direction of the transmission chain. As mentioned above, current studies on COVID-19 have focused more on patient treatment and statistical analysis of COVID-19, and less on the transmission of COVID-19, the mechanism of transmission, and the law of distribution. In particular, no relevant literature has been found to study the thermodynamic imaging principle, COVID-19 transmission mechanism, and diffusion principle of COVID-19 transmission. Therefore, this paper proposes a great innovation in the study of the thermodynamic imaging principle and the evolutionary mechanism of COVID-19 transmission. The structure of the paper is as follows: In Section 2, the paper studies the thermal map imaging principle of COVID-19 transmission, analyzes the thermodynamic imaging characteristics of COVID-19 transmission. In Section 3, a thermodynamic imaging model for COVID-19 transmission is established. In Section 4, the thermodynamic imaging of epidemic transmission in Hubei Province of China was constructed, evaluate and analyze the risk level of 17 cities, and calculation of the superposition density of epidemic situation in Hubei Province. In Section 5 is the conclusions. According to the research problems in this paper, this section mainly introduces the meaning, imaging characteristics, and model architecture of COVID-19 thermal image transmission model, and introduces the thermodynamic image principle and theoretical basis of COVID-19 transmission. The thermodynamic image principle of COVID-19 propagation is mainly used to explain the logic and process of patient aggregation density generation under the COVID-19 background. The model architecture of COVID-19 propagation is mainly used to calculate the thermodynamic density value of regional COVID-19 to provide data basis for the mapping of thermodynamic image. The thermodynamic image shows the high-density region where geographic data activities occur by the density of clustering in the region. Generally, red represents the high-value clustering with high clustering, while blue represents the lowvalue clustering with low density. The darker the color is, the higher the density is, and the more clusters there are. Along with the activity areas and movement characteristics of people, certain geographic space movement transmission rules and aggregation density were presented on the map. By calculating the COVID-19 data within a specific geographical area, and presenting the COVID-19 transmission law and image principle in the form of thermodynamic image, it can help to understand the geographical characteristics of the COVID-19 transmission. The COVID-19 outbreak has been found to be characterized by rapid human-to-human transmission. In other words, in a certain COVID-19 patient, the COVID-19 virus was rapidly transmitted to people in close contact or nearby in a certain friend gathering or in a train, plane, or subway, thus forming the first generation of transmission. Then, these people who were infected with COVID-19 got together with their friends, or they took trains, planes, or subways for business trips and travel, and then they transmit COVID-19 to people who were in close contact with them, or to people who were close to them, and this was the second generation of transmission. A second generation of infected people can cause a third generation of transmission if they continue to infect their close contacts or those around them. If this continues, it will be transmitted from one generation to the next, and the number of infected people will increase, resulting in a serious COVID-19 infection. The geographical location of COVID-19 virus and the transmission mechanism of three generations are shown in Fig. 1 . As shown in Fig. 1 , the COVID-19 is transmitted from the initial gathering place to different regions gradually through person-to-person transmission in one generation, two generations, and successive three generations. After many times of superposition and evolution, COVID-19 statistics such as the cumulative number of confirmed cases, the number of newly confirmed cases, the cumulative number of deaths, and the number of close contacts in a specific region gradually changed, and a certain concentration density provided the basis for the generation of COVID-19 thermodynamic image. Through the thermal image of the COVID-19 situation in a certain period of time, the main transmission path and the continuous transmission area of the COVID-19 situation can be calculated based on the color change of the image. This paper will use different colors to represent the different stages of the diffusion process. The colors are used to represent the number of infected people. By comparing the transmission area map of multiple stages, we can find a process and characteristics of the transmission of the COVID-19, understand the evolution speed, density, and direction of the transmission of the COVID-19, and help us better understand the development of the COVID-19 in a region. The thermal map mainly includes five elements: direction marker, the scale, the legend, the clustering area, the transmission direction of the COVID-19 (www.baidu.com). Among them, (1) Direction marker: it is used to indicate the direction of the thermodynamic image, so as to facilitate the reader to judge the direction and specific position indicated by the thermodynamic image. (2) Scale: it represents the scale of the reduction between the distance on the graph and the actual distance, and scales to grasp the overall and local situation of COVID-19 transmission. (3) Legend: it refers to the symbolic language of the map, such as the text description and geographical names and numbers, mainly streets. (4) Clustering area: the clustering region is divided into high-value clustering region and low-value clustering region. Among them, the part with dark color represents the high-value clustering region, and the part with light color represents the low-value clustering region. The high-value clustering region and the low-value clustering region represent the high-incidence or low-incidence regions, respectively. After the above five elements are described in a COVID-19 area, the thermodynamic image of this area can be made. In the process of making the thermal map, the superposition calculation of the COVID-19 transmission points is needed, and then the superposition density is calculated before the clustering regions of the thermal map are finally generated. Superposition density refers to the density distribution formed by the superposition of infection points from different directions. In the process of applying the thermodynamic imaging principle to study the principle of COVID-19 transmission, the five elements of the thermal diagram, namely orientation criteria, scale, legend, clustering area, and the direction of COVID-19 transmission, each have the following meanings: (i) The orientation criterion is used to represent the direction of the COVID-19 transmission process. (ii) The scale is used to show the proportion between the distance on the map and the actual distance during the transmission of the COVID-19. (iii) The legend is used to represent the transmission of the COVID-19 area, the maps of the text, such as geographical names and numbers of symbolic language. (iv) Cluster regions are used to represent areas of high or low incidence in the course of COVID-19 transmission. Among them, the dark part represents the more serious area, and the light part represents the less serious area; (v) The direction of the transmission of the COVID-19 is used to indicate the main path and direction of the transmission of the COVID-19. The components of COVID-19 thermodynamic image mainly rely on geographical location data, cluster region, and diffusion direction. (1) Geographic location data mainly refers to the map carriers such as street information and regional images formed by GIS on the COVID-19 thermodynamic image images, which are used to determine the exact location and transmission location of virus infection. (2) Clustering region will be formed through calculating the density data of the COVID-19 outbreak of agglomeration areas, used to direct the embodiment of the COVID-19 area and intensity are concentrated, high-value clustering in red, blue means low clustering, refers to the area of high-value clustering gathered more confirmed cases; new cases and close contacts may carry virus-infected patients. The direction and path of COVID-19 evolution and transmission can be analyzed by the arrow direction. (3) The direction of diffusion is according to the arrow logos can determine the direction of the COVID-19 transmission, COVID-19 development to a certain stage, the virus may be from high-value clustering to low clustering transmission in the region and may also be from low clustering area reverse flow to the high-value area transmission, the specific situation shall be calculated according to the actual development data of COVID-19. The direction of COVID-19 transmission and the high-value clustering region and low-value clustering region formed in an area are shown in Fig. 2. In this paper, the thermal image of COVID-19 transmission is generated by density analysis tool. Density analysis can calculate the order of magnitude per unit area based on the point layer or the line map layer and finally simulate the thermal image of smooth surface (www.baidu.com). The dot layer refers to the superimposed graph formed by discrete points of COVID-19 transmission. The line map layer refers to the graph superimposed by a series of curves formed by the continuous points of COVID-19 transmission. In the density analysis, the calculated density map is given different weights, respectively, which is based on the actual situation of the COVID-19 to give the weight of relevant indicators. Finally, the superimposed density map is obtained after weighted calculation based on different density maps. According to the principle of COVID-19 transmission and diffusion, the density map of suspected cases, newly confirmed cases, cumulative confirmed cases, cumulative deaths and close contacts in the region was determined, and the final superposition thermal image was generated by the final weighted calculation. The specific patient indicators in a certain area are updated synchronously within a certain period of time to realize the dynamic law of COVID-19 transmission image. The data of the thermodynamic imaging model of COVID-19 data in this paper are mainly from authoritative official websites. The framework of the imaging model of COVID-19 transmission based on the thermal image principle of COVID-19 is shown in Fig. 3 . Assuming that n indicators A 1 , A 2 , ..., A n (n ≥ 1,n is an integer) of the COVID-19 transmission process in a region are considered, the weight corresponding to each indicator is ω 1 , ω 2 , ..., the same as below), and ∑ n i¼1 ω i ¼ 1, and the superposition density of each indicator is D 1 , D 2 , ..., D n , respectively. Assuming that the total superposition density of the COVID-19 in this region is D, the calculation formula of the total superposition density of the COVID-19 in this region is: In the formula, D i is calculated on the basis of the accumulated data of the corresponding index. Formula (1) is also known as the thermodynamic imaging model of COVID-19 disease transmission. The superposition density of a certain area reflects the risk degree of epidemic spread in the area, which can be used to evaluate the epidemic risk in the area. (D 4 ), and the cumulative number of deaths (D 5 ), with weights of ω 1 , ω 2 , ω 3 , ω 4 , and ω 5 respectively. Therefore, from the point of virus infection and COVID-19 transmission risk, we believe that (1) the greater the number of newly confirmed cases and the accumulated number of confirmed cases, the faster the rate of virus infection and the greater the risk of COVID-19 transmission. Therefore, the newly confirmed cases and the accumulated confirmed cases have the greatest impact on the risk of COVID-19 transmission; they in the five indicators of the weight of ω 1 and ω 2 are the largest, can consider them the same weight, that is, ω 1 = ω 2 . (2) The suspected case has shown signs of infection with a novel coronavirus to a certain extent, but has not yet been confirmed, so the weight of the suspected case in the 5 indicators ω 3 should be the second largest. (3) Because close contacts are likely to be infected, once infected, they will transmit the virus, increasing the risk of transmission of the COVID-19, so it can be considered that close contacts' weight of ω 4 is located in the third place. (4) Since accumulative deaths can be properly sterilized and normalized, the risk of re-transmission caused by infection source is low. Therefore, it can be considered that the weight of cumulative death cases ω 5 accounts for minimum. Thus, we can obtain the following equation and inequality chain: It may be assumed that the weight of new and cumulative confirmed cases is ω 1 = ω 2 = 0.3. The weight of suspected cases is ω 3 = 0.2, the weight of close contacts is ω 4 = 0.15, and the weight of cumulative deaths is ω 5 = 0.05. Then, the superposition density of thermodynamic images in this area is: When the data of D 1 , D 2 , D 3 , D 4 , and D 5 in this region are collected, the superposition density D of the thermal map in this region can be calculated by formula (3). In this paper, the calculated area superposition density was sorted. The region with a large value represented the denser the COVID-19 cluster, while the region with a small value represented the smaller the COVID-19 cluster. According to the size of the numerical value, the regional COVID-19 thermal image was drawn, and the depth of the color represented the size of the superimposed density value, and the main path of the COVID-19 development and diffusion was calculated by the COVID-19 thermal image of the interval date. By comparing the thermal image of the initial date with the final thermal image, the propagation direction curve of the COVID-19 can be drawn. On December 30, 2019, the Health Commission of Wuhan City, Hubei Province, issued the "emergency notice on reporting the treatment situation of pneumonia with unknown causes," which officially announced that patients with pneumonia with unknown causes had been found in the south China seafood market of Wuhan City. Previously, the unexplained pneumonia was detected in Wuhan in November 2019, followed by a number of cases of pneumonia caused by the virus in Wuhan, Hubei Province. Until 21:00 on January 7, 2020, experts got the whole genome sequence of the virus, the virus was isolated from the typical form of coronavirus, and unexplained pneumonia cases the pathogen judgement for this will be coronavirus (https://bigd.big.ac. cn/ncov/release_genome) [18] . To control the transmission of the virus, the Chinese government decided to cut off the transportation between Wuhan and the outside world from January 23, 2020. The outbreak of COVID-19 started in Wuhan, Hubei Province, China, and had been transmitted to the whole country and the world. Hubei Province is the first outbreak place of novel coronavirus. Taking Hubei Province as a case study to study the thermodynamic image principle of COVID-19 propagation is of great significance to the COVID-19 research. On January 23, 2020, the first confirmed cases were reported in the cities of Shiyan and Yichang in Hubei Province, followed by the first confirmed cases in Ezhou City, mainly in Wuhan and Huanggang. Since the reporting data of the COVID-19 in Hubei Province were gradually improved and comprehensive from January 23, 2020, the COVID-19 data in this paper were mainly collected from January 23, 2020. The data collected in this paper are from the official website of the Health Commission of Hubei Province of China. According to the data published on the official website of the Health Commission of Hubei Province of China (http://wjw.hubei.gov.cn/), the total number of confirmed cases, deaths, and close contacts in Hubei Province during the period of March 31, January 23, 2020, is shown in Table 1 . In this paper, the overall situation of COVID-19 in Hubei Province during January 23, 2020 (solstice, March 31) was collected, and the cumulative number of confirmed patients, deaths, and close contacts in Hubei Province during the COVID-19 period was mainly collected, as shown in Table 1 and Fig. 4. In Fig. 4 , we can have a macro understanding of the development of the COVID-19 in Hubei Province. As can be seen from Fig. 4 , February is the period when the number of infected people in Hubei Province of China surges. According to the incubation period of the virus, most of the patients have been infected by the virus in mid-January, and most of the potential patients were detected by screening in February. 549 729 1052 1423 2714 3554 4586 5806 7153 9074 11177 13522 Cumulative deaths 24 39 52 76 100 125 162 204 249 294 350 414 Close contacts 3653 5682 7989 10394 16904 22095 28780 35144 41075 48571 56088 The clinical manifestations of COVID-19 patients are usually fever, fatigue, and dry cough. Most of them will have dyspnea 1 week after infection, and severe patients will rapidly progress to acute respiratory distress syndrome, septic shock, refractory metabolic acidosis, and coagulation dysfunction [19] . According to Table 2 statistics, in January 23, 2020, to March 8, 2020, the infection will be coronavirus and the increase in the number of deaths in the death toll in various areas in Hubei Province increased rapidly. Since the beginning of March, with the deepening of medical research and the intervention of government regulation, the number of deaths has been controlled after patients receive effective treatment, and the growth rate has been slow. Most of the reported deaths were among elderly people with chronic diseases or poor immune function. The capital city of Wuhan, Hubei Province, is the largest inland water, land, and air transportation hub in China and the shipping center of the middle reaches of the Yangtze river. Wuhan's high-speed railway network radiates to most provinces and cities in China, and it is the only city in central China that can directly fly to five continents in the world. The outbreak of COVID-19 has been transmitted from Wuhan to the whole of Hubei Province. According to the data on the official website of Hubei provincial Health Commission, (1) until March 31, 2020, Wuhan became the worst-hit area for COVID-19, and Xiaogan City and Huanggang City had the second and third largest number of confirmed COVID-19 patients, respectively. (2) From the end of January to the middle of February in 2020, the total number of confirmed patients in the province increased rapidly, and the overall trend was explosive. (3) The number of newly diagnosed patients in most areas of Hubei Province, except Wuhan, was effectively controlled around February 20, 2020. Since February 5, Xiaogan City surpassed Huanggang City to become a high-incidence area except Wuhan. The data collected in this paper are from the official website of Hubei Provincial Health Committee of China (http://wjw.hubei.gov.cn/); it is mainly able to obtain the number of newly confirmed cases, cumulative confirmed cases, new death cases, and cumulative death cases in each city of COVID-19 development of 17 cities in Hubei Province, which are Wuhan (WH), Xiaogan (XG), Huanggang (Hg), Jingzhou (QZ), Ezhou (EZ), Suizhou (SZ) In this paper, a number of thermal images are generated by calculating the point density of the region. According to the timeline of the COVID-19 development, this paper mainly selects the COVID-19 data of multiple time nodes in Hubei Province for the generation of thermal images. Through the comparison of the COVID-19 thermal images of multiple time nodes, the evolution law of the COVID-19 transmission is speculated, and the evolution mechanism of the COVID-19 is studied according to the thermal images of the virus transmission. According to the thermodynamic imaging principle, we use red color to represent the high-value clustering region and blue color to represent the low-value clustering region. The image value of clustering region is obtained mainly by calculating the superposition density of the region. The larger the superimposed density value, the more people infected by the COVID-19, and the smaller the superimposed density value, the fewer people infected by the COVID-19. According to stacking density formula (3), this paper calculates the daily stacking density of each region, Hubei Province, to save space is mainly chosen a few more key nodes of time, such as selection of January 23, February 1, February 10, February 19, February 28, March 8, March 17, March 26, and March 31 in 2020, as shown in Table 3 . To calculate the superimposed density of regions, we first need to determine the weights of newly confirmed cases, cumulative confirmed cases, new deaths, and cumulative deaths. According to the previous hypothesis, since the number of newly confirmed cases per day is the focus, the weight of newly confirmed cases should be greater than the weight of cumulative confirmed cases. New deaths per day are the most feared, so they should be given more weight in the four categories. And the weight of cumulative death cases should be a little more than the weight of cumulative confirmed cases, but not to the weight of new death cases, so the weight of cumulative death cases and new confirmed cases can be equally important. Therefore, we assume that the weights of newly confirmed cases, cumulative confirmed cases, new deaths, and cumulative deaths are 0.25, 0.2, 0.3, and 0.25, respectively. According to Table 2 , on January 23, 2020, 70 new cases were confirmed in Wuhan, with a total of 495 confirmed cases, 6 new deaths, and 20 deaths. Then, according to formula (3), the superimposed density value of Wuhan City on that day (reserved integer) is: That is, the superimposed density of Wuhan City is 123. Listing 123 in Table 3 corresponds to the position of the superimposed density value of Wuhan City on January 23. The superimposed density values of other COVID-19 area cities in Hubei Province are also calculated according to formula (4) , and the calculated results are all listed in the corresponding positions in Table 3 . In Table 3 , the cumulative superimposed density value of each city in Hubei Province is calculated to measure the risk level of each city during the rapid transmission of COVID-19. In this paper, we establish the risk level determination rules within the time span as follows: Risklevel ¼ C ≥ 10000 Higher-risk 1000≤ C < 10000 Medium-risk 100 ≤ C < 1000 Low-risk C < 100 where C is the cumulative density of city (or region). From Table 3 , (i) The cumulative density of Wuhan City is 67,977. According to the risk grade determination rule (5), the risk grade of Wuhan City is "highest-risk." (ii) The accumulative superposition density of Xiaogan, Huanggang, Jingzhou, Ezhou, Suizhou, Xiangyang, Huangshi, Yichang, Jingmen, and Xianning were all above 1000 cases and below 10,000 cases. Therefore, according to the risk grade determination rule (5), the risk grade of these 10 cities is "higher-risk." (iii) The accumulative superposition density values of the five cities (states) of Shiyan, Xiantao, Tianmen, Enshi, 1 WH 123 1111 4286 9582 10370 10601 10627 10636 10641 67,977 Highest-risk 2 XG 10 184 566 686 733 735 736 736 736 5122 Higher-risk 3 HG 2 273 502 589 610 613 613 613 613 4428 4 QZ 2 80 144 305 327 328 329 329 329 2173 5 EZ 0 73 181 276 290 292 293 294 294 1993 6 SZ 0 98 234 264 272 273 273 273 273 and Qianjiang are all above 100 cases and below 1000 cases. Therefore, according to the risk grade determination rule (5) , the risk grade of these five cities is "medium-risk." (iv) The cumulative superposition density of Shennongjia is 15. Therefore, according to the risk grade determination rule (5) , the risk grade of Shennongjia is "low-risk." In fact, the design principle of rule (5) are as follows: (1) The cumulative superimposed density of Wuhan in the first place is 67,977, which is far higher than that of Xiaogan City in the second place, which is 5122. Therefore, Wuhan is independently divided into the highest-risk area; (2) Xiaogan City, Huanggang City, Jingzhou City, Ezhou City, Suizhou City, Xiangyang City, Huangshi City, Yichang City, Jingmen City, and Xianning City all have a cumulative superposition density of thousands, so these 10 cities are divided into highrisk areas; (3) Shiyan City, Xiantao City, Tianmen City, Enshi City, and Qianjiang City all have a cumulative superposition density of hundreds, so these five cities are divided into medium-risk areas; (4) The cumulative superposition density of Shennongjia is 15, which is far less than the hundreds number, so Shennongjia is independently divided into lowrisk areas. Figure 6 shows the cumulative density curves of 17 cities (states) in Hubei Province. As can be seen from Fig. 6 , the cumulative density of Wuhan City is far greater than that of other 16 cities in the province. Therefore, Wuhan was indeed a "highest-risk" area during the transmission time of the COVID-19 on January 23, 2020, to March 31, 2020. Assuming C 1 , C 2 , C 3 , C 4 , C 5 , C 5 , C 7 , C 8 , C 9 , C 10 , C 11 , C 12 , C 13 , C 14 , C 15 , C 16 , and C 17 are respectively the cumulative superposition density of Wuhan, Xiaogan, Huanggang, Jingzhou, Ezhou, Suizhou, Xiangyang, Huangshi, Yichang, Jingmen, Xianning, Shiyan, Xiantao, Tianmen, Enshi, Qianjiang, and Shennongjia, according to Table 3 , the cumulative superposition density of these 17 cities can be calculated. The chain relation between the cumulative superposition density value of Wuhan City and the cumulative superposition density value of other 16 cities (states) can be obtained (with 2 decimal places reserved) as follows: Next, we observe the superimposed density growth curve of Wuhan in "highest-risk" areas, as shown in Fig. 7 . As can be seen from Fig. 7 According to the calculation formula of stacking density and the available data, this paper calculated the outbreak of Hubei Province during the development of new confirmed cases of individual key node density value, the cumulative density value, new cases of the death cases of density value, the density of total deaths, with four density values of the weighted superposition to calculate the last outbreak of thermodynamic image overlay density values. According to the superimposed density, the COVID-19 thermodynamic images of nine key time points in Hubei Province were drawn. Through the thermal image of the COVID-19, we can find the law and direction of the evolution and transmission of the COVID-19, and put forward the corresponding risk control methods and suggestions. In the early stage of the outbreak, there were risks and viruses in other cities except Wuhan, but they were not found. According to the results of COVID-19 thermal map imaging, the fermentation and evolution of the COVID-19 in Hubei Province are divided into four stages: incubation period, initial period, outbreak period, and stationary period; the thermodynamic imaging characteristics of the four stages are shown in the following figures. ( Fig. 9 . It can be seen from Fig. 9 that the thermodynamic imaging color is gradually deepened and the imaging range is gradually expanded in this stage. The number of confirmed infections ranges from small to large Fig. 8 The initial period of COVID-19 in Hubei Province controlled, the number of newly diagnosed patients has decreased significantly for several consecutive days, the trend of the COVID-19 development is generally favorable, and the number of patients discharged from hospital has increased substantially. Since mid-to-late March, the number of newly confirmed cases in several areas of Hubei Province, except Wuhan City, has reached 0, and the outbreak has been further controlled, which is related to the Chinese government's strong quarantine and COVID-19 control measures. At this stage, the total number of confirmed cases in Hubei Province has stabilized at about 68,000, but the imported cases will pose new challenges for the Chinese government. This paper mainly focused on the analysis of Hubei Province of China, but since mid-to-late March in Hubei Province, the newly added cases in many places except Wuhan City were 0, so the thermodynamic image of this paper was only mapped to March 31, 2020. Thermodynamic image during stable period in Hubei Province is shown in Fig. 11 . It can be seen from Fig. 11 that in this stage, the thermodynamic imaging color becomes the deepest, the imaging range is expanded to the pole, and the imaging area is all over the province. This indicates that the novel coronavirus pneumonia peak has arrived, the epidemic development has reached saturation point, and the turning point has begun to appear. The whole epidemic development is gradually entering a stable period. Based on the regional COVID-19 development data, this paper constructed the thermodynamic image principle model of regional COVID-19 situation, and drew thermodynamic image of COVID-19 based on the actual COVID-19 data, which provided a new reference direction for the study of COVID-19. Taking Hubei Province, where the outbreak occurred, as the research object, the thermodynamic image mechanism of the outbreak development, transmission, and diffusion in Hubei Province was analyzed, and the aggregation density, evolution direction, and evolution speed of the outbreak fermentation, transmission, and diffusion in Hubei Province were obtained from the visual perspective. Then, a statistical model was established to study the superimposed density of the COVID-19 development in Hubei Province, and the cumulative superimposed density of the COVID-19 development in each city was used to evaluate the risk level. We come to the following conclusions: (1) In the early stage of the COVID-19, except Wuhan, other cities in Hubei Province had COVID-19 risks and viruses, but no confirmed cases were detected and reported in time. Thermodynamic imaging is a kind of point-like distribution state, which presents a trend of random walk and outward transmission. (2) In the incubation period of COVID-19, Xiaogan is the first report confirmed cases outside Wuhan, Huanggang, such as Wuhan area, mid-to-late January of Shiyan, Yichang, Wuhan distant equidistance area also gradually began to report the first confirmed case, according to the virus incubation period for 14 days, most of the patients in early January has been infected with the virus and the virus to be transmitted to other areas. Thermodynamic imaging color gradually deepened, imaging range gradually expanded; (3) In the outbreak period of COVID-19, the number of infected people in Wuhan was mainly concentrated in Wuchang district, Qiaokou district, and Jiangan district. The areas of Xiaogan, Hanchuan, and Anlu were the hardest hit areas. The number of infected people in Huangzhou and Hongan of Huanggang also ranked in the forefront of the city. The number of confirmed COVID-19 cases in Hubei Province has exceeded 67,000. On March 8, 2020, the total number of confirmed COVID-19 cases in Hubei Province was 67,743, an increase of 67,194 from January 23, 2020. Among them, the number of confirmed COVID-19 cases in Wuhan accounted for 74% of the total number of infected people in the province, making it the hardest hit area. Thermodynamic imaging color rapidly deepened, imaging range rapidly expanded, imaging area rapidly increased. (4) In the stable period of the COVID-19, that is, since midto-late March, the number of newly confirmed patients in several areas of Hubei Province, except Wuhan City, was 0, and the COVID-19 was further controlled, which was related to the strong quarantine and COVID-19 control measures of the Chinese government. Thermodynamic imaging color became deepest, imaging range expanded to the poles, imaging areas throughout the province. and Qianjiang, the cumulative superimposed density of the COVID-19 was above 100 and below 1000, and the risk level of these five cities was "medium-risk Jingzhou 31.28 times, 34.11 times of Ezhou, Suizhou 34.68 times, 37.68 times of Xiangyang, 45.40 times of Huangshi, 48.24 times of Yichang References 1. Update on covid-19 outbreak as of 24:00 on 1 May (2020) National Health Commission of China COVID-19) situation report -102 (2020) World Health Organization Feng Z (2020) Early transmission dynamics in A novel coronavirus from patients with pneumonia in China Mathematical model of infection kinetics and its analysis for covid-19, SARS and MERS Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study Epidemiologic and clinical characteristics of novel coronavirus infections involving 13 patients outside Wuhan, China Heat map generation for predictive safety planning: preventing struck-by and near miss interactions between workers-on-foot and construction equipment Heat map visualisation of fire incidents based on transformed sigmoid risk model Analysing the spatial-temporal characteristics of bus travel demand using the heat map Using multi-source big data to understand the factors affecting urban park use in Wuhan Safe mobile robot navigation in human-centered environments using a heat map-based path planner Predicting pathological response to neoadjuvant chemotherapy in breast cancer patients based on imbalanced clinical data Evidence-based personal applications of medical computing models in risk factors of cardiovascular disease for the middle-aged and elderly Assessing the utility of a smart thermometer and mobile application as a surveillance tool for influenza and influenza-like illness Framework for infectious disease analysis: a comprehensive and integrative multimodeling approach to disease prediction and management ncovr) (2020) National Biological Information Center of China Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Epidemic situation of COVID -19 in Hubei province on Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations The number of confirmed infections ranges from small to large Fig. 11 Thermodynamic image during stable period in Hubei Province The research method in this paper provides a scientific analysis method for the transmission principle, evolutionary mechanism, and urban risk assessment of COVID-19 transmission.Funding The paper was supported by the National Natural Science Foundation of China grant number 91646112 and the Key Programmes of the Science and Technology Department of Guangdong Province grant number 2016A020224001.