Research article

Mathematical modeling and optimal control strategies of Buruli ulcer in possum mammals

  • Received: 17 February 2021 Accepted: 13 May 2021 Published: 29 June 2021
  • MSC : 92J15, 34C08

  • Buruli is a neglected tropical disease that can now be found in many countries including developed countries like Australia. It is a skin disorder that usually occurs in the arms and legs. The disease has been identified in a number of mammals, particularly possum. Adequate eradication and control programs are needed to minimize infection and its spread to less developed countries before it becomes an epidemic. In this work, a SIR type possum epidemic model is proposed. The properties of the model are thoroughly studied and obtained its stability results. We determine the stability of the model at its fixed points and show that the model is locally and globally asymptomatically stable. The stability of the disease-free case is shown for R0<1 and the endemic case is examined for R0>1. We further extend the model using the control variables and obtain an optimal control system and using the optimal control theory to characterize the necessary condition for controlling the spread of Buruli ulcer (BU). The model results are plotted to determine the best strategies for disease elimination. Numerical simulation has shown that the useful strategy consists in implementing all suggested controls.

    Citation: Muhammad Altaf Khan, E. Bonyah, Yi-Xia Li, Taseer Muhammad, K. O. Okosun. Mathematical modeling and optimal control strategies of Buruli ulcer in possum mammals[J]. AIMS Mathematics, 2021, 6(9): 9859-9881. doi: 10.3934/math.2021572

    Related Papers:

    [1] Ahmed Alshehri, Saif Ullah . Optimal control analysis of Monkeypox disease with the impact of environmental transmission. AIMS Mathematics, 2023, 8(7): 16926-16960. doi: 10.3934/math.2023865
    [2] Suprawee Lertnaweephorn, Usa Wannasigha Humphries, Amir Khan . Stability analysis and optimal control for leaf brown spot disease of rice. AIMS Mathematics, 2023, 8(4): 9602-9623. doi: 10.3934/math.2023485
    [3] Shinta A. Rahmayani, Dipo Aldila, Bevina D. Handari . Cost-effectiveness analysis on measles transmission with vaccination and treatment intervention. AIMS Mathematics, 2021, 6(11): 12491-12527. doi: 10.3934/math.2021721
    [4] Meroua Medjoudja, Mohammed El hadi Mezabia, Muhammad Bilal Riaz, Ahmed Boudaoui, Saif Ullah, Fuad A. Awwad . A novel computational fractional modeling approach for the global dynamics and optimal control strategies in mitigating Marburg infection. AIMS Mathematics, 2024, 9(5): 13159-13194. doi: 10.3934/math.2024642
    [5] Mingzhan Huang, Shouzong Liu, Ying Zhang . Mathematical modeling and analysis of biological control strategy of aphid population. AIMS Mathematics, 2022, 7(4): 6876-6897. doi: 10.3934/math.2022382
    [6] Youming Guo, Tingting Li . Dynamics and optimal control of an online game addiction model with considering family education. AIMS Mathematics, 2022, 7(3): 3745-3770. doi: 10.3934/math.2022208
    [7] Huihui Zhong, Weijian Wen, Jianjun Fan, Weijun Yang . Reinforcement learning-based adaptive tracking control for flexible-joint robotic manipulators. AIMS Mathematics, 2024, 9(10): 27330-27360. doi: 10.3934/math.20241328
    [8] Sayed Saber, Azza M. Alghamdi, Ghada A. Ahmed, Khulud M. Alshehri . Mathematical Modelling and optimal control of pneumonia disease in sheep and goats in Al-Baha region with cost-effective strategies. AIMS Mathematics, 2022, 7(7): 12011-12049. doi: 10.3934/math.2022669
    [9] Houssine Zine, Abderrahim El Adraoui, Delfim F. M. Torres . Mathematical analysis, forecasting and optimal control of HIV/AIDS spatiotemporal transmission with a reaction diffusion SICA model. AIMS Mathematics, 2022, 7(9): 16519-16535. doi: 10.3934/math.2022904
    [10] Kai Zhang, Yunpeng Ji, Qiuwei Pan, Yumei Wei, Yong Ye, Hua Liu . Sensitivity analysis and optimal treatment control for a mathematical model of Human Papillomavirus infection. AIMS Mathematics, 2020, 5(3): 2646-2670. doi: 10.3934/math.2020172
  • Buruli is a neglected tropical disease that can now be found in many countries including developed countries like Australia. It is a skin disorder that usually occurs in the arms and legs. The disease has been identified in a number of mammals, particularly possum. Adequate eradication and control programs are needed to minimize infection and its spread to less developed countries before it becomes an epidemic. In this work, a SIR type possum epidemic model is proposed. The properties of the model are thoroughly studied and obtained its stability results. We determine the stability of the model at its fixed points and show that the model is locally and globally asymptomatically stable. The stability of the disease-free case is shown for R0<1 and the endemic case is examined for R0>1. We further extend the model using the control variables and obtain an optimal control system and using the optimal control theory to characterize the necessary condition for controlling the spread of Buruli ulcer (BU). The model results are plotted to determine the best strategies for disease elimination. Numerical simulation has shown that the useful strategy consists in implementing all suggested controls.



    In this paper, we consider only undirected, connected, and simple graphs. We use V(G) and E(G) to denote the vertex set and the edge set of a graph G, respectively.

    An edge-coloring of a graph G is an assignment of colors to its edges so that no two edges incident to the same vertex receive the same color. An edge-coloring σ of G using k colors (k edge-coloring) is then a partition of the edge set E(G) into k disjoint matchings and it can be written as σ=(M1,M2,,Mk), where every Mi is a matching of G. The chromatic index of G, denoted by χ(G), is the least k for which G has a k edge-coloring. The Vizing Theorem states that χ(G)=Δ(G) or χ(G)=Δ(G)+1. Graphs with χ(G)=Δ(G) are said to be Class 1; graphs with χ(G)=Δ(G)+1 are said to be Class 2. It is NP-complete to determine whether a graph is Class 1 [3]. The classification problem is extremely difficult even for regular graphs. For this problem, Chetwynd and Hilton [1] proposed the following conjecture:

    Conjecture 1.1 (1-Factorization Conjecture). Let G be a k-regular graph with n vertices, n even. If kn2 then χ(G)=k.

    In this paper, we consider the generalized lexicographic product of graphs, which is defined as follows [10]: Let G be a graph with V(G)={t1,t2,,tn}, n2, and hn=(Hi)i{1,2,,n} be a sequence of vertex-disjoint graphs with V(Hi)={(ti,y1),(ti,y2),,(ti,ymi)}, mi1. The generalized lexicographic product G[hn] of G and hn=(Hi)i{1,2,,n} is a simple graph with vertex set ni=1V(Hi), in which (ti,yp) is adjacent to (tj,yq) if and only if either ti=tj and (ti,yp)(ti,yq)E(Hi) or titjE(G). A generalized lexicographic product is also called an expansion or composition (see[11]).

    By Vi, i=1,2,,n we will denote the vertex set of graph Hi in G[hn], and call the sets V1,V2,,Vn the partition sets of G[hn]. If HiH for i=1,2,,n, then G[hn]=G[H], where G[H] is the lexicographic product of two graphs G and H. For example, the join H1+H2 of vertex-disjoint graphs H1 and H2 is K2[h2], where h2=(Hi)i{1,2}. In addition, Turán graphs Tr1(n) (see[2]) are complete (r1)-partite graphs with nr1 vertices whose partition sets differ in size by at most 1, that is, Tr1(n) are Kr1[hr1], where hr1=(Hi)i{1,2,,r1} is a sequence of vertex-disjoint empty graphs with nr1 vertices or nr1 vertices.

    De Simone and Picinin de Mello [9] gave the following sufficient conditions for a join graph to be Class 1:

    Theorem 1.1. Let G=H1+H2 be a join graph with |V(H1)||V(H2)|. If Δ(H1)>Δ(H2), then G is Class 1.

    Theorem 1.2. Let G=H1+H2 be a join graph with Δ(H1)=Δ(H2). If both H1 and H2 are Class 1, or if H1 is a subgraph of H2, or if both H1 and H2 are disjoint unions of cliques, then G is Class 1.

    Theorem 1.3. Every regular join graph G=H1+H2 with Δ(H1)=Δ(H2) is Class 1.

    De Simone and Galluccio [8] showed that 1-Factorization Conjecture is true for graphs that are join of two graphs:

    Theorem 1.4. Every regular join graph with even order is Class1.

    De Simone and Galluccio [6,7] extended the above result, and proved the following conclusions:

    Theorem 1.5. Every even graph that is the join of two regular graphs is Class 1.

    Mohar [5] and Jaradat [4] gave the following sufficient conditions for a lexicographic product of graphs to be Class 1:

    Theorem 1.6. Let G and H be two graphs. If G is Class 1, then G[H] is Class 1.

    Theorem 1.7. Let G and H be two graphs. If χ(H)=Δ(H) and H is of even order, then χ(G[H])=Δ(G[H]).

    By Theorems 1.6 and 1.7, it is easy to see that for any two regular graphs G and H, if 1-Factorization Conjecture is true for the graph G, or 1-Factorization Conjecture is true for the graph H and Δ(G)|V(G)|12, then this conjecture is also true for graphs that are lexicographic product G[H] of G and H.

    In this paper, our goal is to find sufficient conditions for a generalized lexicographic product G[hn] of G and hn=(Hi)i{1,2,,n} to be Class 1, where G is a graph with n vertices and all graphs in hn have the same number of vertices.

    The following lemma will be used later:

    Lemma 1. (Jaradat[8])Let G and H be two graphs such that χ(H)=Δ(H). Then χ(G×H)=Δ(G×H), where G×H denotes the direct product of graphs G and H.

    In Section 2, we shall present two decompositions of the edge set in the generalized lexicographic product of graphs.

    Let G be a graph with V(G)={t1,t2,,tn}, n2, and hn=(Hi)i{1,2,,n} be a sequence of vertex-disjoint graphs with V(Hi)={(ti,y1),(ti,y2),,(ti,ym)}, m1. Let G=G[hn], and let Uj={(t1,yj),(t2,yj),,(tn,yj)}, where j=1,2,,m. Then Gj=G[Uj]G for each j=1,2,,m. We will provide two decompositions of the generalized lexicographic product G.

    We first consider the case where G is a Class 1 graph and provide a decomposition of G with respect to a matching of G. If there exists a matching M in G such that χ(GM)=Δ(G)1 and any distinct vertices with the maximum degree of G are saturated by distinct edges of M, then G is said to be Subclass 1; otherwise, G is said to be Subclass 2. For example, every Class 1 graph in which no two vertices of maximum degree are adjacent is Subclass 1, and every regular Class 1 graph is Subclass 2.

    For every matching M of G such that χ(GM)=Δ(G)1, let GM denote the subgraph of G induced by M. Moreover, let

    GM=GM[¯Km](tiVΔHi),G1=G1[¯Km](tiV(G)VΔHi),

    where VΔ denotes the set of vertices with the maximum degree in G, ¯Km denotes an empty graph with m vertices, and G1=GM. Then G is the union of edge-disjoint graphs GM and G1, that is,

    G=GMG1. (2.1)

    Figure 1 shows a Subclass 1 graph G and the decomposition of G=G[h5] with respect to a matching M in G, where hn=(Hi)i{1,2,,5} is a sequence of vertex-disjoint graphs, each with m vertices. Figure 2 shows a Subclass 2 graph G and the decomposition of G=G[h6] with respect to a matching M in G, where hn=(Hi)i{1,2,,6} is a sequence of vertex-disjoint graphs, each with m vertices.

    Figure 1.  The decomposition of G with respect to a matching M in a Subclass 1 graph G.
    Figure 2.  The decomposition of G with respect to a matching M in a Subclass 2 graph G.

    We now provide another decomposition of G. Let

    G2=(mj=1Gj)(ni=1Hi),G3=G×Km.

    Then the graph G is the union of edge-disjoint graphs G2 and G3, that is,

    G=G2G3. (2.2)

    In Section 3, we shall study sufficient conditions for G to be Class 1 using the above two decompositions of G.

    Through the decomposition of G in formula 2.1, we can make the following observation:

    Observation 3.1. Let G be a Class 1 graph. If there exists a matching M in G such that χ(GM)=Δ(G)1 and the corresponding GM is Class 1, then G is also Class 1.

    Proof. Let M be a matching of G such that χ(GM)=Δ(G)1 and the corresponding GM is Class 1. It is easy to see that Δ(GM)=max{Δ(Hi)|tiVΔ}+m and Δ(G1)=(Δ(G)1)m. Note that Δ(G)=max{Δ(Hi)|tiVΔ}+Δ(G)m. Hence, Δ(G)=Δ(GM)+Δ(G1). It follows that if both GM and G1 are Class 1, then G is Class 1. Thus, we only need to verify that χ(G1)=(Δ(G)1)m.

    Since χ(G1)=Δ(G)1, it follows that we can color the edges of G1[¯Km] with (Δ(G)1)m colors such that for each positive integer i, tiV(G)VΔ, there are at least m colors are missing at all vertices in Vi. Note that G1=G1[¯Km](tiV(G)VΔHi). Hence, we can extend the (Δ(G)1)m edge-coloring of G1[¯Km] to all the edges of tiV(G)VΔHi, so that χ(G1)=(Δ(G)1)m.

    Theorem 3.1. If G is Subclass 1, then there exists a matching M in G such that χ(GM)=Δ(G)1 and the corresponding GM is Class 1.

    Proof. Let M be a matching of G such that χ(GM)=Δ(G)1 and any distinct vertices with the maximum degree in G are saturated by distinct edges of M. Note that each connected component of GM is either a join of a graph on m vertices and an empty graph on m vertices, or a balanced complete bipartite graph on 2m vertices. It is easy to see that these connected components are all Class 1. Thus GM is Class 1.

    An instant corollary of Theorem 3.1 and Observation 3.1 is:

    Corollary 3.1. If G is Subclass 1, then G is Class 1.

    Theorem 3.2. Let G be a Subclass 2 graph, and let M be a matching of G such that χ(GM)=Δ(G)1. For every pair of vertices tp and tq of maximum degree in G which are saturated by the same edge of M, if one of the following five conditions holds:

    (i) both Hp and Hq are Class 1;

    (ii) Hp is a subgraph of Hq;

    (iii) both Hp and Hq are disjoint unions of cliques;

    (iv) Δ(Hp)Δ(Hq);

    (v) join graph Hp+Hq is regular;

    then GM is Class 1.

    Proof. Since G is Subclass 2, every connected component of GM can be denoted by Hp+Hq, or Hp+¯Km, or Hq+¯Km, or ¯Km+¯Km, where ¯Km denotes an empty graph on m vertices, and ¯Km+¯Km denotes the join of two vertex-disjoint empty graphs on m vertices. Note that Hp+¯Km, Hq+¯Km and ¯Km+¯Km are all Class 1. Hence, it is only necessary to prove that: if one of the conditions (ⅰ)–(ⅴ) holds, then Hp+Hq is Class 1.

    Assume that one of the conditions (ⅰ)–(ⅳ) holds. Since |V(Hp)|=|V(Hq)|=m, it follows from Theorems 1.1 and 1.2 that Hp+Hq is Class 1.

    Assume that (ⅴ) holds. Since Hp+Hq is regular and |V(Hp)|=|V(Hq)|, Hp+Hq is Class 1 by Theorem 1.3 or Theorem 1.4.

    An instant corollary of Theorem 3.2 and Observation 3.1 is:

    Corollary 3.2. Let G be a Subclass 2 graph, and let M be a matching of G such that χ(GM)=Δ(G)1. For every pair of vertices tp and tq of maximum degree in G which are saturated by the same edge of M, if one of five conditions of Theorem 3.2 holds, then G is Class 1.

    Note that the join graph H1+H2 corresponds to P2[h2], where P2 is a Class 1 graph. By Corollaries 3.1 and 3.2, we can effortlessly generalize the results of Theorem 1.5 in the case where |V(H1)|=|V(H2)| to the generalized lexicographic product, yielding the following theorem:

    Theorem 3.3. If G is Class 1 and all graphs in hn are regular, then G is Class 1.

    In addition, by Corollaries 3.1 and 3.2, we can directly obtain Theorem 1.6. Furthermore, one easy consequence of Corollary 3.2 is the following result, which is similar to Theorem 1.2.

    Theorem 3.4. Let G=H1+H2 be a join graph with |V(H1)|=|V(H2)|. If G is regular, or if both H1 and H2 are Class 1, or if H1 is a subgraph of H2, or if both H1 and H2 are disjoint unions of cliques, then G is Class 1.

    Through the decomposition of G in formula 2.2, we can make the following observation:

    Observation 3.2. Suppose that all graphs in hn have the same maximum degree. If both G2 and G3 are Class 1, then G is Class 1.

    Proof. Let Δ(Hi)=ΔH, where i=1,2,,n. It is easy to see that Δ(G2)=Δ(G)+ΔH, Δ(G3)=Δ(G)(m1) and Δ(G)=Δ(G)m+ΔH. Hence, Δ(G)=Δ(G2)+Δ(G3). Since χ(G2)=Δ(G2), and since χ(G3)=Δ(G3), it follows that χ(G)=Δ(G), that is, G is Class 1.

    By Observation 3.2, we can obtain the following theorem:

    Theorem 3.5. Suppose that all graphs in hn are Class 1 graphs with the same maximum degree. If m is even, then G is Class 1.

    Proof. Let Δ(Hi)=ΔH, where i=1,2,,n. We can first color the edges of each subgraph Gj of G2 with Δ(G)+1 colors 1,2,,Δ(G)+1 such that edges which are corresponding to the same edge of G receive the same color. Since we use Δ(G)+1 colors, it follows that each vertex (ti,yj) of Hi misses at least one color ci in {1,2,,Δ(G)+1}, where i=1,2,,n. Hence, we can color the edges of each subgraph Hi of G2 with the color ci and an additional ΔH1 new colors. Thus, χ(G2)Δ(G)+ΔH=Δ(G2), that is, G2 is Class 1. On the other hand, since G3=G×Km, and since m is even, G3 is Class 1 according to Lemma 1.1. Therefore, G is also Class 1.

    An instant corollary of Theorem 3.5 is:

    Corollary 3.3. If all graphs in hn are regular Class 1 graphs with the same maximum degree, then G is Class 1.

    By applying Theorem 3.5, we can derive Theorem 1.7. Furthermore, by utilizing Theorem 3.3 and Corollary 3.3, we can formulate the following theorem:

    Theorem 3.6. Suppose that G=G[hn] is regular. If G is Class 1, or each graph of hn is Class 1, then G is Class 1.

    Proof. Since G is regular, and since all graphs in hn have the same number of vertices, it follows that all graphs in hn are regular graphs with the same maximum degree. If G is Class 1, then G is Class 1 by Theorem 3.3. If each graph of hn is Class 1, then G is Class 1 according to Corollary 3.3.

    By applying Theorem 3.6, take G=P2, we can derive the result of theorem 1.4 in the case where |V(H1)|=|V(H2)|. In addition, the graph G in Theorem 3.6 does not necessarily satisfy the condition "Δ(G)|V(G)|2" of 1-Factorization Conjecture. For instance, suppose that G is a k-regular bipartite graph with n vertices, and hn=(Hi)i{1,2,,n} represents a sequence of vertex-disjoint cubic graphs, each with m vertices, where m4. Clearly, G is Class 1 according to Theorem 3.6, however, we have Δ(G)=km+3<nm2=|V(G)|2 when k<n21.

    It is easy to see that if G is regular and all graphs in hn have the same number of vertices, then the inequality Δ(G)|V(G)|2 can be derived from the inequality Δ(G)|V(G)|2. By Theorem 3.6, if the 1-factorial conjecture holds for G, then the conjecture holds for G.

    Finally, in a generalized lexicographic product Kp[hp], we consider the case where |V(Kp[hp])| is even and every graph in the sequence hp=(Hi)i{1,2,,p} has |V(Kp[hp])|p vertices or |V(Kp[hp])|p vertices. Using Theorem 1.5, we can derive the following theorem:

    Theorem 3.7. Let G=Kp[hp] and let hp=(Hi)i{1,2,,p} be a sequence of vertex-disjoint k-regular graphs such that every graph in hp has |V(G)|p vertices or |V(G)|p vertices, where p2 and k0. If |V(G)| is even, then G is Class 1.

    Proof. Let n=|V(G)|, and let V1,V2,,Vp be the partition sets of G. If np=np, it follows that G is an even graph that is the join of two regular graphs, then G is Class 1 according to Theorem 1.5. If npnp, then we may assume that |Vi|=np for i=1,2,,r and |Vi|=np for i=r+1,r+2,,p, where 1rp1. Let

    G1=G[ri=1Vi],G2=G[pi=r+1Vi].

    Clearly, both G1 and G2 are regular graphs, and G=G1+G2. By Theorem 1.5, G is Class 1.

    In Theorem 3, 7, by letting k=0 and p=r1, then we can directly derive the following corollary:

    Corollary 3.4. All Turán graphs on an even number of vertices are Class 1.

    In this paper, for a generalized lexicographic product G[hn] of a graph G with n vertices and a sequence hn=(Hi)i{1,2,,n} of vertex-disjoint graphs with m vertices, we obtain the following sufficient conditions for G[hn] to be Class 1: (ⅰ) G is Subclass 1; (ⅱ) G is Class 1 and all graphs in hn are regular; (ⅲ) all graphs in hn are Class 1 graphs with the same maximum degree, and m is even; (ⅳ) G[hn] is regular and either G or each graph in hn is Class 1.

    In addition, for a generalized lexicographic product G=Kp[hp] of a complete graph Kp on p vertices and a sequence hp=(Hi)i{1,2,,p} of vertex-disjoint k-regular graphs whose partition sets differ in size by at most 1, we prove that G is Class 1 if G has an even number of vertices.

    The authors declare that they have not used Artificial Intelligence (AI) tools in the creation of this article.

    Fund projects: Mathematics, Gansu Province Key Discipline (11080318). Applied Mathematics National Minority Committee Key Discipline (11080327), and Support for innovation team of operations research and Cybernetics in Northwest University for Nationalities.

    All authors declare no conflicts of interest in this paper.



    [1] E. Bonyah, I. Dontwi, F. Nyabadza, A Theoretical Model for the Transmission Dynamics of the Buruli Ulcer with Saturated Treatment, Comput. Math. Meth. Med., 2014 (2014), 1–14.
    [2] E. Bonyah, K. Badu, S. Kwes, Optimal control application to an Ebola model, Asian Pac. J. Trop. Biom., 6 (2016), 283–289. doi: 10.1016/j.apjtb.2016.01.012
    [3] F. Nyabadza, E. Bonyah, On the transmission dynamics of Buruli ulcer in Ghana: Insights through a mathematical model, BMC Res. Notes, 8 (2015), 1–15. doi: 10.1186/1756-0500-8-1
    [4] E. Bonyah, I. Dontwi, F. Nyabadza, Optimal Control Applied to the Spread of Buruli Uclcer Disease, Am. J. Comput. Appl. Math., 4 (2014), 61–76.
    [5] K. Blayneh, Y. Cao, H. Kwon, Optimal control of vector-borne diseases: Treatment and Prevention, Disc. Cont. Dyn. Sys. Ser. B, 11 (2009), 587–611.
    [6] O. D. Makinde, K. O. Okosun, Impact of chemo-therapy on optimal control of malaria disease with infected immigrants, BioSystems, 104 (2011), 32–41. doi: 10.1016/j.biosystems.2010.12.010
    [7] E. Bonyah, M. A. Khan, K. O. Okosun, S. Islam, A theoretical model for Zika virus transmission, Plos one, 12 (2017), e0185540. doi: 10.1371/journal.pone.0172713
    [8] K.O. Ouifki, N. Marcus, Optimal control analysis of a malaria disease transmission model that includes treatment and vaccination with waning immunity, BioSystems, 106 (2011), 136–145. doi: 10.1016/j.biosystems.2011.07.006
    [9] S. Ullah, M. A. Khan, Modeling the impact of non-pharmaceutical interventions on the dynamics of novel coronavirus with optimal control analysis with a case study, Chaos Soliton. Fract., 139 (2020), 110075. doi: 10.1016/j.chaos.2020.110075
    [10] E. O. Alzahrani, W. Ahmad, M. A. Khan, M. J. Malebary, Optimal Control Strategies of Zika Virus Model with Mutant, Commun. Nonl. Sci. Num. Simul., 93 (2021), 105532. doi: 10.1016/j.cnsns.2020.105532
    [11] M. A. Khan, L. Ahmed, P. K. Mandal, R. Smith, M. Haque, Modelling the dynamics of Pine Wilt Disease with asymptomatic carriers and optimal control, Sci. Rep., 10 (2020), 1–15. doi: 10.1038/s41598-019-56847-4
    [12] S. Ullah, O. Ullah, M. A. Khan, T. Gul, Optimal control analysis of tuberculosis (TB) with vaccination and treatment, Eur. Phy. Jour. Plus, 135 (2020), 1–27. doi: 10.1140/epjp/s13360-019-00059-2
    [13] M. A. Khan, S. A. A. Shah, S. Ullah, K. O. Okosun, Optimal control analysis of the effect of treatment, isolation and vaccination on hepatitis B virus, J. Biol. Syst., 28 (2020), 351–376. doi: 10.1142/S0218339020400057
    [14] J. P. Lasalle, Stability theroy for difference equations, J. K. Hale (Ed.) Studies in Ordinary Differential Equations, Washington DC: Math. Assoc. of America, 1977.
    [15] E. Bonyah, K. O Okosun, Mathematical modeling of Zika virus, Asi. Paci. Jour. Trop. Dis., 6 (2016), 673–679. doi: 10.1016/S2222-1808(16)61108-8
    [16] W. H. Fleming, R. W. Rishel, Deterministic and stochastic optimal control, Springer Verlag, New York (1975).
    [17] L. S. Pontryagin, The mathematical theory of optimal processes, Wiley, New York, 1962.
    [18] K. Asiedu, F. Portaels, Mycobacterium ulcerans infection, World Health Organisation, Global Buruli Ulcer Initiative, 2000.
    [19] A. Bolliger, B. R. V. Forbes, W. B. Kirkland, Transmission of a recently isolated mycobacterium to phalangers (Trichosurusvulpecula), Science, 12 (1950), 146–147.
    [20] J. Hayman, Postulated epidemiology of Mycobacterium ulcerans infection, Inter. J. Epi., 20 (1991), 1093–1098. doi: 10.1093/ije/20.4.1093
    [21] G. Birkhoff, G. C. Rota, Ordinary differential equations, Ginn, Boston, 1982.
    [22] P. V. D. Driessche, J. Watmough, Reproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission, Math. Biosci., 180 (2002), 29–48. doi: 10.1016/S0025-5564(02)00108-6
    [23] U. D. Purwati, F. Riyudha, H. Tasman, Optimal control of a discrete age-structured model for tuberculosis transmission, Heliyon, 6 (2020), e03030. doi: 10.1016/j.heliyon.2019.e03030
    [24] H. Tasman, Optimal control of HIV resistance and tuberculosis co-infection using treatment intervention, Asi. Pac. Jour. Trop. Dis., 7 (2017), 366–373. doi: 10.12980/apjtd.7.2017D6-400
    [25] Fatmawati, D. Utami, M. U. I. Purwati, C. Alfiniyah, Y. Prihartini, The dynamics of tuberculosis transmission with optimal control analysis in Indonesia, Commun. Math. Biol. Neur., 2020 (2020), 1–17.
    [26] L. Marsollier, R. Robert, J. Aubry, J. P. Saint Andre, H. Kouakou, P. Legras, et al., Aquatic insects as a vector for Mycobacterium ulcerans, Appl. Environ. Microbiol., 68 (2002), 4623–4628. doi: 10.1128/AEM.68.9.4623-4628.2002
    [27] J. A. Hayman, J. Hibble, Mycobacterium ulcerans in wild animals, Rev. sci. tech. Off. Int. Epiz., 20 (2001), 252–264.
    [28] D. L. Lukes, Differential equations: Classical to Controlled, Mathematics in Science and Engineering, Academic Press, New York, 1982.
    [29] J. A. M. Fyfe, C. J. Lavender, K. A. Handasyde, A. R. Legione, C. R. O'Brien, T. P. Stinear, et al., A Major Role for Mammals in the Ecology of Mycobacterium ulcerans, PLoS. Negl. Trop. Dis., 4 (2010), e791. doi: 10.1371/journal.pntd.0000791
    [30] L. Marsollier, R. Robert, J. Aubry, J. Saint Andre, H. Kouakou, et al., Aquatic insects as a vector for Mycobacterium ulcerans, Appl. Environ. Microbiol., 68 (2002), 4623–4628. doi: 10.1128/AEM.68.9.4623-4628.2002
    [31] H. R. Williamson, M. E. Benbow, K. D. Nguyen, D. C. Beachboard, R. K. Kimbirauskas, Distribution of Mycobacterium ulcerans in Buruli Ulcer Endemic and NonEndemic Aquatic Sites in Ghana, PLoS. Negl. Trop. Dis., 2 (2008), 1–15.
    [32] F. Portaels, W. M. Meyers, A. Ablordey, A. G. Castro, K. Chemlal, et al., First Cultivation and Characterization of Mycobacterium ulcerans from the Environment, PLoS. Negl. Trop. Dis., 2 (2008), 40–55.
  • This article has been cited by:

    1. S. Olaniyi, M. Mukamuri, K.O. Okosun, O.A. Adepoju, Mathematical analysis of a social hierarchy-structured model for malaria transmission dynamics, 2022, 34, 22113797, 104991, 10.1016/j.rinp.2021.104991
    2. S. Olaniyi, M. Mukamuri, Kazeem Okosun, O.A. Adepoju, Mathematical Analysis of a Social Hierarchy-Structured Model for Malaria Transmission Dynamics, 2021, 1556-5068, 10.2139/ssrn.3932213
    3. Rubin Fandio, Hamadjam Abboubakar, Henri Paul Ekobena Fouda, Anoop Kumar, Kottakkaran Sooppy Nisar, Mathematical modelling and projection of Buruli ulcer transmission dynamics using classical and fractional derivatives: A case study of Cameroon, 2023, 8, 26668181, 100589, 10.1016/j.padiff.2023.100589
  • Reader Comments
  • © 2021 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(3374) PDF downloads(213) Cited by(3)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog