Research article Special Issues

An effective classifier based on convolutional neural network and regularized extreme learning machine

  • Received: 24 April 2019 Accepted: 25 August 2019 Published: 18 September 2019
  • An effective classifier combining convolutional neural network and regularized extreme learning machine (called as CNN-RELM) is presented in this paper. Firstly, CNN-RELM trains the convolutional neural network (CNN) using the gradient descent method until the learning target accuracy reaches. Then the fully connected layer of CNN is replaced by regularized extreme learning machine (RELM) optimized by genetic algorithm and the rest layers of the CNN remain unchanged. The experiments on different face databases are given to evaluate the performance of CNN-RELM. The experimental results show that CNN-RELM is a feasible classifier and it outperforms CNN and RELM. Due to the uniting of CNN and RELM, CNN-RELM have the advantages of CNN and RELM and it is easier to learn and faster in testing.

    Citation: Chunmei He, Hongyu Kang, Tong Yao, Xiaorui Li. An effective classifier based on convolutional neural network and regularized extreme learning machine[J]. Mathematical Biosciences and Engineering, 2019, 16(6): 8309-8321. doi: 10.3934/mbe.2019420

    Related Papers:

    [1] Weike Zhou, Aili Wang, Fan Xia, Yanni Xiao, Sanyi Tang . Effects of media reporting on mitigating spread of COVID-19 in the early phase of the outbreak. Mathematical Biosciences and Engineering, 2020, 17(3): 2693-2707. doi: 10.3934/mbe.2020147
    [2] Hamdy M. Youssef, Najat A. Alghamdi, Magdy A. Ezzat, Alaa A. El-Bary, Ahmed M. Shawky . A new dynamical modeling SEIR with global analysis applied to the real data of spreading COVID-19 in Saudi Arabia. Mathematical Biosciences and Engineering, 2020, 17(6): 7018-7044. doi: 10.3934/mbe.2020362
    [3] Haonan Zhong, Wendi Wang . Mathematical analysis for COVID-19 resurgence in the contaminated environment. Mathematical Biosciences and Engineering, 2020, 17(6): 6909-6927. doi: 10.3934/mbe.2020357
    [4] Fang Wang, Lianying Cao, Xiaoji Song . Mathematical modeling of mutated COVID-19 transmission with quarantine, isolation and vaccination. Mathematical Biosciences and Engineering, 2022, 19(8): 8035-8056. doi: 10.3934/mbe.2022376
    [5] Pannathon Kreabkhontho, Watchara Teparos, Thitiya Theparod . Potential for eliminating COVID-19 in Thailand through third-dose vaccination: A modeling approach. Mathematical Biosciences and Engineering, 2024, 21(8): 6807-6828. doi: 10.3934/mbe.2024298
    [6] Lin Feng, Ziren Chen, Harold A. Lay Jr., Khaled Furati, Abdul Khaliq . Data driven time-varying SEIR-LSTM/GRU algorithms to track the spread of COVID-19. Mathematical Biosciences and Engineering, 2022, 19(9): 8935-8962. doi: 10.3934/mbe.2022415
    [7] Yujie Sheng, Jing-An Cui, Songbai Guo . The modeling and analysis of the COVID-19 pandemic with vaccination and isolation: a case study of Italy. Mathematical Biosciences and Engineering, 2023, 20(3): 5966-5992. doi: 10.3934/mbe.2023258
    [8] Ahmed Alshehri, Saif Ullah . A numerical study of COVID-19 epidemic model with vaccination and diffusion. Mathematical Biosciences and Engineering, 2023, 20(3): 4643-4672. doi: 10.3934/mbe.2023215
    [9] Dipo Aldila, Meksianis Z. Ndii, Brenda M. Samiadji . Optimal control on COVID-19 eradication program in Indonesia under the effect of community awareness. Mathematical Biosciences and Engineering, 2020, 17(6): 6355-6389. doi: 10.3934/mbe.2020335
    [10] Tianfang Hou, Guijie Lan, Sanling Yuan, Tonghua Zhang . Threshold dynamics of a stochastic SIHR epidemic model of COVID-19 with general population-size dependent contact rate. Mathematical Biosciences and Engineering, 2022, 19(4): 4217-4236. doi: 10.3934/mbe.2022195
  • An effective classifier combining convolutional neural network and regularized extreme learning machine (called as CNN-RELM) is presented in this paper. Firstly, CNN-RELM trains the convolutional neural network (CNN) using the gradient descent method until the learning target accuracy reaches. Then the fully connected layer of CNN is replaced by regularized extreme learning machine (RELM) optimized by genetic algorithm and the rest layers of the CNN remain unchanged. The experiments on different face databases are given to evaluate the performance of CNN-RELM. The experimental results show that CNN-RELM is a feasible classifier and it outperforms CNN and RELM. Due to the uniting of CNN and RELM, CNN-RELM have the advantages of CNN and RELM and it is easier to learn and faster in testing.


    Coronavirus belongs to the Coronaviridae family in phylogeny. It was first identified as a pathogenic microorganism of human disease in the mid-1960s. The name comes from the fact that the ratchet protein on the surface of the virus envelope protrudes like a crown under an electron microscope. Before 2019, six known coronaviruses types have been discovered: α genus of 229E, β genus of OC43, HKU1, Middle East respiratory syndrome coronavirus (MERS-COV) and severe acute respiratory syndrome (SARS-COV) [1,2,3]. In December 2019, an unidentified pneumonia was found in Wuhan, Hubei province, China. The responsible virus was later confirmed as a new coronavirus [4]. The National Board of Health name the disease as a novel coronavirus pneumonia (NCP). The World Health Organization (WHO) temporarily named the virus as the novel coronavirus 2019 (2019-nCov), and the disease as coronavirus disease 2019 (COVID-19) on 11 February. At the same time, the International Committee for the classification of viruses named the virus as severe acute respiratory syndrome coronavirus-2 (SARS-COV2) [5]. The disease spread rapidly from Wuhan to all parts of the country and overseas. Till March 12, 2020 at 5:44 p.m, a total of 80,981 people had been confirmed infected in China, and the total death toll had reached 3173. New cases of COVID-19 have emerged in 109 countries overseas (including Japan, South Korea and so on), with 38,620 confirmed cases and 1465 deaths get rid of China from the WHO [6].

    Anhui, a neighbouring province of Hubei, is vulnerable to the disease. Since the first confirmed case reported on January 10, 2020 in Anhui, a total of 990 cases have been confirmed and 6 deaths have been reported by March 12. Figure 1 shows the heat map of the total number of confirmed cases in China on February 6. The epidemic poses a challenge to the prevention and control of the epidemic in Anhui.

    Figure 1.  Heat map of the total number of confirmed cases in China on February 6.

    In the face of the severe epidemic, the Chinese government promptly adopted a series of policies to mitigate the epidemic. The lockdown of the epicenter Wuhan has been implemented since January 23, 2020. On February 6, Anhui province started to adopt the centralized quarantine measures for close contacts of cases. These policies effectively prevented the spread of the disease in Anhui. This is the first work to use mathematical modeling techniques combined with observed data to evaluate the control effectiveness in Anhui. As the outbreak of COVID-19 now became a pandemic, the lesson learnt in Anhui, China is surely valuable for mitigation in other localities. Data-driven mathematical modelling plays an important role in epidemic mitigation, in preparedness for future epidemic and in the evaluation of control effectiveness. The basic reproduction numbers (number of secondary cases may be caused by a typical primary case in a wholly susceptible population, R0) is one of the key epidemiological parameters in quantifying the rate of disease transmission, the size of the outbreak and the needed control effort [7,8]. Many data-driven modelling works in different regions have been done [9,10,11]. In this work, we focused on the epidemic in Anhui based on daily morbidity data which reflect the epidemic more accurately and promptly than laboratory confirmed data. Different from previous studies [8,9,10,11,12,13,14,15], we adopted relatively simple models with flexible transmission rates and model structures.

    The interrupted time series analysis is a quasi-experimental research design. By collecting and analyzing relevant data of the outcome indicators measured at multiple time points before and after the implementation of an intervention, the change trend of the outcome indicators before and after the implementation of the intervention is compared to evaluate the impact of the intervention on the outcome indicators [16,17]. We obtained the daily morbidity data from January 10 to February 11, 2020 from the Anhui Province Center of Disease Control and Prevention. During this period, two interventions occurred, Wuhan lockdown on January 23 (i.e., no imported cases from Wuhan to Anhui thereafter), and centralized quarantine management measures for close contacts implemented since February 6. We adopted an interrupted time series model to evaluate the effects of these two interventions.

    We define the interrupted time series model as follows: Yt=β0+β1+timet+β2intervention1t+β3timeafterintervention1t+β4+intervention2t+β5timeafterintervention2t+et, where Yt is the daily morbidity index value, β0 is the baseline value (also initial value at t=0); β1 is the baseline slope value (before any intervention); β2 refers to the contribution of the first intervention on the baseline value; β3 is the contribution of the first intervention on the baseline slope; That is, β1+β3 is the trend after the implementation of the first intervention; similarly, β4,5 are contribution of the second intervention on the baseline value and baseline slope; β4,5 replaced β1,2 after the implementation of the second intervention. Thus, β1+β5 is the trend of daily morbidity after the second intervention.

    As shown in Figure 2, the model prediction (the black curve) is divided into three sections, before January 23 (phase Ⅰ), between January 23 and February 6 (phase Ⅱ) and after February 6 (phase Ⅲ). The slopes of in phases Ⅰ-Ⅲ are β1, β1+β3, β1+β5, respectively. The instantaneous and lasting effects of the two interventions are clearly demonstrated.We fit the model with Eviews and find the r-squared R2=0.92 and the Durbin-Watson statistic DW=2.32, which indicates high fitting performance of the model. Parameter estimates are listed in the Table 1.

    Figure 2.  Trend of morbidity number of COVID-19 based on piecewise regression model.
    Table 1.  Fitting parameter values of interrupted time series model.
    Parameter Estimate p-value
    β0 7.076923 0.0778
    β1 2.923077 0.0000
    β2 38.29670 0.0000
    β3 -6.057143 0.0000
    β4 -4.742857 0.5012
    β5 -0.323077 0.8442

     | Show Table
    DownLoad: CSV

    According to Table 1, the linear growth trend of the morbidity number during phase Ⅰ was β1=2.9 (pvalue<1e5). The morbidity number had a sudden lift β2=38 (pvalue<1e5) on January 23, 2020. After that the slope of the morbidity number decreased by β3=6.057 (pvalue<1e5). These results imply that right before the lockdown of Wuhan a large number of imported cases from Wuhan or other Hubei cities arrived into Anhui. The lockdown effect was substantial as the morbidity data showed an immediate down trend. However, the effect of the second intervention (after February 6) was noticeable but no significant. As shown in Figure 2, the daily number of cases continued to decrease after February 6. To sum up, it can be seen that the lockdown measures in Wuhan as well as the centralized quarantine management measures in Anhui had a great impact against the spread of COVID-19 in Anhui.

    According to the Anhui Provincial Center for Disease Control and Prevention, the onset of the first case in Anhui was on January 10, 2020 and the case was confirmed on January 22, 2020. There is a 12 days confirmation delay. According to the report of the National Health Commission, both pre-symptomatic patients and symptomatic patients may be contagious. Recent studies showed that the mean serial interval (time delay between the symptom onset of a primary case and a second case) is short, and could be shorter than the incubation period (time delay from infection to symptom onset). Thus pre-symptomatic transmission (one day or two prior symptom onset) is fully justified [9,10].

    Our interrupted time series analysis showed that both Wuhan lockdown and Anhui centralized quarantine management had an impact on the daily morbidity number in Anhui. During the initial phase of the COVID-19 outbreak, the rate of exposure could be high due to lack of awareness and protection. Subsequently, the relevant announcement and measures issued by the government attracted public's attention and enhanced level of self-protection, and strict quarantine measures were adopted. The rate of exposure gradually declined, and the form of epidemic transmission (main transmitter) may have changed accordingly. We consider these factors and formulate our mechanistic models for the three phases.

    Phase Ⅰ (prior to January 23, 2020): During this period public awareness of the severity of the disease was low and cases (especially those of mild symptom) might be treated as common cold. Both pre-symptomatic cases (especially right before symptom onset) and symptomatic cases were infectious [8]. Considering the infectivity before and after the symptom onset might be different, we divided the infected people into two categories: E and I in phase Ⅰ, and allowed the infectivity of class E to be less than that of class I. However, the mobility of class I might be lower than that of class E, thus the infectivity level of E might be comparable to I. The population N can be divided into four categories: susceptible (S), exposed and pre-symptomatic population (E), symptomatic population (I), and recovered population (R). The total population N equals S+E+I+R. The transmission dynamic model reads:

    {dSdt=βS(kE+I)N,dEdt=βS(kE+I)NqE,dIdt=qEγI,dRdt=γI. (2.1)

    where, β is the transmission rate of disease, q is the rate at which exposed individual showing symptom, {k controls the infectiousness of exposed pre-symptomatic individuals relative to symptomatic individuals, γ is the cure rate at which symptomatic individuals moving to recovered class. We ignore COVID-19 induced death since there were no deaths from diseases in the early stage and few deaths in the later stage in Anhui.

    Phase Ⅱ (between January 23 and February 6, 2020): public began to realize the seriousness of the epidemic after Wuhan lockdown on January 23 and lockdown in other cities in Hubei shortly after. The country also issued the epidemic situation announcement, and began to take quarantine measures, symptomatic cases went to the hospital to see a doctor and would be quarantined before confirmation. Therefore, after January 23, only pre-symptomatic cases in E class can effectively contact susceptibles to spread the disease. And there was also a decrease in the rate of exposure after January 23. Therefore, on the basis of model (2.1), we incorporate the policy intervention factors, and change the exposure rate to α1β. We assume all symptomatic cases were quarantined in quarantined class (Q), and σ is the cure rate at which quarantined cases moving to recovered class. Then SEQR model (2.2) reads:

    {dSdt=α1βSEN,dEdt=α1βSENqE,dQdt=qEσQ,dRdt=σQ. (2.2)

    Phase Ⅲ (after February 6, 2020): The centralized quarantined management measures for close contacts was adopted on February 6 in Anhui and resulted in a decline in the chance of contact among the population. Therefore, after February 6, the rate of exposure rate was changed to α2β. The model is the same as (2.2), except for α2 replacing α1.

    We adopted MCMC method implemented in R package to estimate parameters in models (2.1 and 2.2). We fit I(t) or Q(t) to the observed morbidity data. Note that our morbidity are daily current symptomatic cases. We obtained β=0.4029, q=0.41 (per day), k=0.78 and γ=0.182 (per day) in the first stage. Since during the first phase, individuals in E and I were not quarantined, both of them may spread the disease. We obtained the basic reproduction number during the phase 1: R0=kβq+βγ=2.9764. We did not distinguish imported cases and local cases and treat all cases as local. A substantial proportion of cases in phase Ⅰ would be imported cases from Wuhan or other cities in Hubei. If the proportion of imported cases over the time during the phase Ⅰ was relatively constant, our method is still relatively reasonable since the mortality is still a proxy of the local epidemic. Given that most of the infections occurred among family members, we presume the ratio of imported versus local case were relatively stable during phase Ⅰ, and gradually vanished later. The actual and predicted values of the morbidity number during phase Ⅰ are compared in Figure 3.

    Figure 3.  Morbidity number and actual value in phase Ⅰ (actual value in red circle and predicted value in blue).

    We used MCMC to fit model 2 to the morbidity number between January 23 and February 6, 2020. We obtained α1β=0.208, q=0.24, σ=0.04. Using generation matrix method, the basic reproduction number was estimated R0=α1βq=0.8667. This reduction in R0 was due to lockdown effects, holiday effects, and public being vigilant. The actual and predicted values of the morbidity number during phase Ⅱ are compared Figure 4.

    Figure 4.  Morbidity number and actual value in phase Ⅱ (actual value in red circle and predicted value in blue).

    Similarly, we fitted model (2.2) (with α2 replacing α1) to the morbidity number after February 6, 2020. We obtained α2β=0.1311, q=0.229, σ=0.051. The basic reproduction number is R0=α2βq=0.5725. This reduction in R0 was due to further strengthened quarantine measure in Auhui.

    The comparison between actual value and predicted value of daily morbidity in phase Ⅲ is show in Figure 5. Under the policies of Wuhan lockdown and centralized quarantined management in Anhui, the morbidity number has been reduced and sharply decreased.

    Figure 5.  Morbidity number in Phase Ⅲ (actual value is red circle and predicted value is blue).

    The parameter descriptions and estimates are listed in Table 2.

    Table 2.  Parameter description and estimate in Anhui province 2019–2020.
    Parameter Description Values Methods
    β phase Ⅰ transmission rate per capita per day 0.4029 MCMC
    q rate of symptom onset of pre-symptomatic exposed cases 0.229-0.41 Data and MCMC
    α1β phase Ⅱ transmission rate per capita per day 0.208 MCMC
    σ The cure rate 0.04-0.051 Data and MCMC
    α2β phase Ⅲ transmission rate per capita per day 0.1311 MCMC

     | Show Table
    DownLoad: CSV

    Finally, the real data during the whole epidemic period compared to the predicted value of the three-stages model is shown in Figures 6 & 7.

    Figure 6.  Comparison between the morbidity number per day and the observed data.
    Figure 7.  Comparison between the number of new cases per day in the three stages and the observed data.

    In order to view the effect of relevant measures taken by the government on epidemic prevention and control more intuitively, we considered a hypothetical scenario that neither Wuhan lockdown nor centralized quarantine measures were in place and compared the morbidity number with and without government measures in Figure 8.

    Figure 8.  Comparison of the morbidity number per day in Anhui province with or without control measures.

    Furthermore, we considered a second hypothetical scenario when the Wuhan lockdown was in place but no centralized quarantine measures in Anhui. The comparison of morbidity number of patients in Anhui is shown in Figure 9. It can be seen from the graph that the number of new cases of COVID-19 would continue for a long time, rather than decrease as seen now.

    Figure 9.  Comparison of the COVID-19 cases in Anhui province with Wuhan lockdown but no centralized quarantine.

    To sum up, it can be seen that Wuhan lockdown and centralized quarantine measure in Anhui had played a timely and effective role in the prevention and control of the epidemic.

    Based on the results of time series analysis, and the different measures taken by the government as time nodes, this paper establishes a transmission dynamics model which is divided into three phases. It can be seen from the analysis results of each stage that under the gradual implementation of various government measures, the basic reproduction number of the transmission of COVID-19 in Anhui decreased gradually, from 2.9764 to 0.8667 and then to 0.5725. These result of simulations indicate that the epidemic in Anhui has been controlled and is now in a critical period of transmission interruption. At the same time, this also further showed that when a major epidemic occurs, the timely implementation of national policies plays an important role in the control of the epidemic.

    In addition, from the analysis of the data and model results, it can be concluded that the Wuhan lockdown in the second phase reduced the rate of exposure of transmission substantially in Anhui province by 48.37%. A centralized quarantine measure adopted in Anhui reduced the rate of exposure by 67.46% or 36.97%) compared with the rate of exposure rate in phase Ⅰ or phase Ⅱ, respectively. Therefore, both Wuhan lockdown and Anhui timely centralized quarantine measures have greatly reduced the speed of the transmission. The concentration quarantine measures in Anhui caused a substantial decrease in the rate of exposure, thus should be considered timely and effective. From the perspective of public health, this work further illustrated the impact of national and provincial policies in public health.

    This research is supported by National Natural Science Foundation of China (11401002, 11771001), Natural Science Fund for Colleges and Universities in Anhui Province (KJ2018A0029), Teaching Research Project of Anhui University (ZLTS2016065) and Anhui Provincial Department of Science and Technology, Anhui Provincial Health Commission Emergency Research Project "Epidemiological and clinical characteristics of new coronavirus pneumonia" (No. 202004a07020002).

    The authors declare there is no conflicts of interest.



    [1] Y. Bengio, Learning deep architectures for AI, Found. Trends Mach. Learn., 1 (2009), 1-71.
    [2] F. Schroff, D. Kalenichenko and J. Philbin, FaceNet: A Unified Embedding for Face Recognition and Clustering, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), United States of America, 2015, 815-823. Available from: https://arxiv.org/pdf/1503.03832.pdf.
    [3] G. Huang, Z. Liu, L. van der Maaten, et al., Densely Connected Convolutional Networks, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), United States of America, 2017, 4700-4708. Available from: https://arxiv.org/abs/1608.06993.
    [4] Z. Q. Zhao, P. Zheng, S. T. Xu, et al., Object Detection with Deep Learning: A Review, IEEE Trans. Neural Networks Learn. Syst., (2019), 1-21.
    [5] Y. L. Cun, L. D. Jackel, B. Boser, et al., Handwritten digit recognition: Applications of neural network chips and automatic learning, IEEE Commun. Mag., 11 (1989), 41-46.
    [6] I. J. Goodfellow, J. P. Abadie, M. Mirza, et al., Generative Adversarial Networks, Advances in neural information processing systems, 2014, 2672-2680. Available from: https://arxiv.org/abs/1406.2661.
    [7] G. E. Hinton, S. Osindero and Y. W. Teh, A fast learning algorithm for deep belief nets, Neural Comput., 18 (2006), 1527-1554.
    [8] Y. Z. Xu, X. J. Yao, X. Li, et al, Target detection in high resolution remote sensing images based on full convolution network, Bull. Surv. Mapp., 1 (2018), 77-82.
    [9] M. S. Hasan, An application of pre-trained CNN for image classification, 2017 20th International Conference of Computer and Information Technology (ICCIT), 2017. Available from: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8281779.
    [10] X. Qin, Y. Zhou, Z. He, et al., A Faster R-CNN Based Method for Comic Characters Face Detection, 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), 2017, 1074-1080. Available from: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8270109.
    [11] R. Ranjan, V. M. Patel and R. Chellappa, HyperFace: A Deep Multi-Task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition, IEEE Trans. Pattern Anal. Mach. Intell., 41 (2019), 121-135.
    [12] S. Xie and H. Hu, Facial Expression Recognition Using Hierarchical Features with Deep Comprehensive Multipatches Aggregation Convolutional Neural Networks, IEEE Trans. Multimedia, 21 (2019), 211-220.
    [13] G. Chen, C. Li, W. Wei, et al., Fully convolutional neural network with augmented atrous spatial pyramid pool and fully connected fusion path for high resolution remote sensing image segmentation, Appl. Sci., 9 (2019), 1816.
    [14] A. Kulikajevas, R. Maskeliūnas, R. Damaševičius, et al., Reconstruction of 3D Object Shape Using Hybrid Modular Neural Network Architecture Trained on 3D Models from ShapeNetCore Dataset, Sensors, 19 (2019), 1553.
    [15] G. B. Huang, Q. Y. Zhu and C. K. Siew, Extreme learning machine: A new learning scheme of feedforward neural networks, Neural Networks, 2 (2004), 985-990.
    [16] C. He, F. Xu, Y. Liu, et al., A fast kernel extreme learning machine based on conjugate gradient, Network Comput. Neural Syst., 29 (2018), 70-80.
    [17] G. B. Huang, X. Ding and H. Zhou, Optimization method based extreme learning machine for classification, Neurocomputing, 74 (2010), 155-163.
    [18] W. Zou, F. Yao, B. Zhang, et al., Back Propagation Convex Extreme Learning Machine, Proc. ELM, 9 (2016), 259-272.
    [19] C. He, Y. Liu, T. Yao, et al., A fast learning algorithm based on extreme learning machine for regular fuzzy neural network, J. Intell. Fuzzy Syst., 36 (2019), 3263-3269.
    [20] W. Deng, Q. Zheng and L. Chen, Regularized Extreme Learning Machine, 2009 IEEE Symposium on Computational Intelligence and Data Mining, 2009, 389-395. Available from: https://ieeexplore.ieee.org/abstract/document/4938676.
    [21] W. Y. Deng, Q. H. Zheng, L. Chen, et al., Study on extreme learning method of neural network, Chin. J. Comput., 33 (2010), 279-287.
  • This article has been cited by:

    1. Guo-Rong Xing, Ming-Tao Li, Li Li, Gui-Quan Sun, The Impact of Population Migration on the Spread of COVID-19: A Case Study of Guangdong Province and Hunan Province in China, 2020, 8, 2296-424X, 10.3389/fphy.2020.587483
    2. Wei Yang, Modeling COVID-19 Pandemic with Hierarchical Quarantine and Time Delay, 2021, 2153-0785, 10.1007/s13235-021-00382-3
    3. Bootan Rahman, Evar Sadraddin, Annamaria Porreca, The basic reproduction number of SARS‐CoV‐2 in Wuhan is about to die out, how about the rest of the World?, 2020, 30, 1052-9276, 10.1002/rmv.2111
    4. Roman Cherniha, Vasyl’ Davydovych, A Mathematical Model for the COVID-19 Outbreak and Its Applications, 2020, 12, 2073-8994, 990, 10.3390/sym12060990
    5. Ghulam Hussain, Tahir Khan, Amir Khan, Mustafa Inc, Gul Zaman, Kottakkaran Sooppy Nisar, Ali Akgül, Modeling the dynamics of novel coronavirus (COVID-19) via stochastic epidemic model, 2021, 60, 11100168, 4121, 10.1016/j.aej.2021.02.036
    6. Haitao Song, Feng Li, Zhongwei Jia, Zhen Jin, Shengqiang Liu, Using traveller-derived cases in Henan Province to quantify the spread of COVID-19 in Wuhan, China, 2020, 101, 0924-090X, 1821, 10.1007/s11071-020-05859-1
    7. Tingzhe Sun, Yan Wang, Modeling COVID-19 epidemic in Heilongjiang province, China, 2020, 138, 09600779, 109949, 10.1016/j.chaos.2020.109949
    8. Dhiraj Kumar Das, Anupam Khatua, T.K. Kar, Soovoojeet Jana, The effectiveness of contact tracing in mitigating COVID-19 outbreak: A model-based analysis in the context of India, 2021, 00963003, 126207, 10.1016/j.amc.2021.126207
    9. Wei Wei, Bowen Duan, Min Zuo, Quanmin Zhu, An extended state observer based U-model control of the COVID-19, 2021, 00190578, 10.1016/j.isatra.2021.02.039
    10. Prasantha Bharathi Dhandapani, Dumitru Baleanu, Jayakumar Thippan, Vinoth Sivakumar, On stiff, fuzzy IRD-14 day average transmission model of COVID-19 pandemic disease, 2020, 7, 2375-1495, 208, 10.3934/bioeng.2020018
    11. Tahir Khan, Gul Zaman, Youssef El Khatib, Modeling the dynamics of novel coronavirus (COVID-19) via stochastic epidemic model, 2021, 22113797, 104004, 10.1016/j.rinp.2021.104004
    12. Haitao Song, Zhongwei Jia, Zhen Jin, Shengqiang Liu, Estimation of COVID-19 outbreak size in Harbin, China, 2021, 0924-090X, 10.1007/s11071-021-06406-2
    13. Orwa Al-Abdulla, Agneta Kallström, Camilo Valderrama, Jussi Kauhanen, Simulation of the Progression of the COVID-19 Outbreak in Northwest Syria Using a Basic and Adjusted SIR Model, 2022, 2, 2813-0227, 44, 10.3390/zoonoticdis2020006
    14. S. A. Alblowi, M. El Sayed, M. A. El Safty, Decision Making Based on Fuzzy Soft Sets and Its Application in COVID-19, 2021, 30, 1079-8587, 961, 10.32604/iasc.2021.018242
    15. Caitriona Murphy, Jessica Y. Wong, Benjamin J. Cowling, Nonpharmaceutical interventions for managing SARS-CoV-2, 2023, Publish Ahead of Print, 1070-5287, 10.1097/MCP.0000000000000949
    16. Wenjing Zhao, Lin Zhang, Junling Wang, Lili Wang, How has academia responded to the urgent needs created by COVID-19? A multi-level global, regional and national analysis, 2022, 0165-5515, 016555152210846, 10.1177/01655515221084646
    17. Juan Luis Fernández-Martínez, Zulima Fernández-Muñiz, Ana Cernea, Andrzej Kloczkowski, Ezequiel López-Rubio, Predictive Mathematical Models of the Short-Term and Long-Term Growth of the COVID-19 Pandemic, 2021, 2021, 1748-6718, 1, 10.1155/2021/5556433
    18. M. El Sayed, M. A. El Safty, M. K. El-Bably, Topological approach for decision-making of COVID-19 infection via a nano-topology model, 2021, 6, 2473-6988, 7872, 10.3934/math.2021457
    19. ROMAN M. CHERNIHA, VASYL V. DAVYDOVYCH, A reaction–diffusion system with cross-diffusion: Lie symmetry, exact solutions and their applications in the pandemic modelling, 2022, 33, 0956-7925, 785, 10.1017/S095679252100022X
    20. Junaid Iqbal Khan, Jebran Khan, Furqan Ali, Farman Ullah, Jamshid Bacha, Sungchang Lee, Artificial Intelligence and Internet of Things (AI-IoT) Technologies in Response to COVID-19 Pandemic: A Systematic Review, 2022, 10, 2169-3536, 62613, 10.1109/ACCESS.2022.3181605
    21. Haitao Song, Fang Liu, Feng Li, Xiaochun Cao, Hao Wang, Zhongwei Jia, Huaiping Zhu, Michael Y. Li, Wei Lin, Hong Yang, Jianghong Hu, Zhen Jin, Modeling the second outbreak of COVID-19 with isolation and contact tracing, 2022, 27, 1531-3492, 5757, 10.3934/dcdsb.2021294
    22. Haiyin Li, Yan Wu, Toshikazu Kuniya, Dynamics of SCIR Modeling for COVID-19 with Immigration, 2022, 2022, 1099-0526, 1, 10.1155/2022/9182830
    23. Bootan Rahman, Sarbaz H. A. Khoshnaw, Grace O. Agaba, Fahad Al Basir, How Containment Can Effectively Suppress the Outbreak of COVID-19: A Mathematical Modeling, 2021, 10, 2075-1680, 204, 10.3390/axioms10030204
    24. Ting Zeng, Zhidong Teng, Ramziya Rifhat, Xiaodong Wang, Lei Wang, Kai Wang, Analysis and simulation of a stochastic COVID-19 model with large-scale nucleic acid detection and isolation measures: A case study of the outbreak in Urumqi, China in August, 2022, 2023, 24680427, 10.1016/j.idm.2023.03.004
    25. S. Sankaranarayanan, An approach to statistical analysis-using the average transmission model of Covid 19, 2020, 8, 23193786, 2158, 10.26637/MJM0804/0141
    26. Alaa Falih Mahdi, Hussein K. Asker, A Sensitivity Analysis of the PSITPS Epidemic Model’s Parameters for COVID-19, 2024, 11, 2518-0010, 79, 10.31642/JoKMC/2018/110210
    27. Akash Pal, Garima Jain, Ishita Roy, Sumit Sharma, 2023, Chapter 37, 978-981-19-5190-9, 453, 10.1007/978-981-19-5191-6_37
  • Reader Comments
  • © 2019 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(8617) PDF downloads(717) Cited by(22)

Figures and Tables

Figures(8)  /  Tables(3)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog