Abstract
We are concerned by imminent future problems caused by biological dangers, here we think of a way to solve them. One of them is analyzing endemic models, for this we make a study supported by Computer Algebra Systems (CAS) and Mechanized Reasoning (MR). Also we show the advantages of the use of “CAS” and “MR” to obtain in that case, an epidemic threshold theorem. We prove a previously obtained theorem for SnIR endemic model. Moreover using “CAS+MR” we obtain a new epidemic threshold theorem for the SnImR epidemic model and for the staged progressive SImR model. Finally we discuss the relevance of the theorems and some future applications.
Access provided by Autonomous University of Puebla. Download chapter PDF
Similar content being viewed by others
1 Introduction
At the moment, we are at the edge of a possible biological problem. Some people say that the nineteenth century was the century of chemistry, the twentieth was the century of physics, and they say that the twenty-first will be the century of biology. If we think, the advances in the biological field in the recent years have been incredible, and like the physics and its atomic bomb, with biology could create global epidemics diseases. Also the climate change could produce a new virus better than the existing virus, creating an atmosphere of panic. For these reasons and others, we think in a solution using mathematical models with computer algebra and mechanized reasoning. Specifically we consider the SIR (Susceptible-Infective-Removed) model, with differential susceptibility and multiple kinds of infected individuals. The objective is to derive three epidemic threshold theorems by using the algorithm MKNW given in [1] and a little bit of mechanized reasoning.
Briefly the MKNW runs on: Initially we have a system of ordinary non-lineal differential equations S, whose coefficients are polynomial. We start setting all derivates to zero for finding equilibrium; we solve the system finding the equilibrium point T. Then we compute the Jacobian Jb for the system S and replace T in S. We compute the eigenvalues for Jb; from the eigenvalues we obtain the stability conditions when each eigenvalue is less than zero. Finally we obtain the reproductive number for the system S in the particular cases. Using deductive reasoning we obtain some theorems based on the particular cases.
The MKNW algorithm is not sufficient to prove the threshold theorems that will be considered here and for this reason, it is necessary to use some form of mechanized reasoning, specifically some strategy of mechanized induction.
The threshold theorem that we probe in Section 2 was originally presented in [2] using only pen and paper and human intelligence. A first contribution of this paper is a mechanized derivation of such theorem using CAS.
The threshold theorem to be proved in Section 3 is original and some particular cases of this theorem were previously considered via CAS in [3, 4] and without CAS in [5].
The threshold theorem to be proved in Section 4 is original and similar models were before considered without CAS in [6].
2 CA And MR Applied to the SNIR Epidemic Model
We introduce the system for the SnIR epidemic model, which has n groups of susceptible individuals and which is described by the next system of Eq. [2]:
we define the rate of infection as:
and we define p i as follow:
This is a system with (n + 2) equations and each previous constant is defined like it is shown:
-
μ is the natural death rate.
-
γ is the rate at which infectives are removed or become immune.
-
δ is the disease-induced mortality rate for the infectives.
-
ε is the disease-induced mortality rate for removed individuals.
-
αi is the susceptibility of susceptible individuals.
-
β is the infectious rate of infected individuals.
-
η is the average number of contacts per individual.
Each function or group is defined as follow:
-
X i(t) are the n groups of susceptible in the time equal t
-
Y(t) is the group of infectives in the time equal t.
-
Z(t) is the group of removed in the time equal t.
2.1 The Standard SIR Model
As a particular case we analyze the standard SIR model [7] which has just one group of susceptible and is described in the next equation system:
In the infection-free equilibrium there is no variation in time. So the derivates are canceled and the solution for the previous system, it’s given by:
After, we generate the Jacobian matrix for the equations system:
we substitute the solution (5) in the Jacobian:
Now, we find the eigenvalues for the previous matrix:
and its corresponding stability condition is:
this can be rewritten as:
using the next expression:
Finally we find the basic reproduction number which it represents the condition of equilibrium:
2.2 The S2IR Model
As another particular case we analyze the S2IR model where there are two groups of susceptible and the equations for this system are:
Initially we find the infection-free equilibrium solution for the previous system:
Equally we generate the Jacobian matrix for the equations system and substituting the infection-free equilibrium point in the Jacobian:
we find the eigenvalues for the previous Jacobian,
and the corresponding stability condition is:
this can be rewritten as:
also it can be written using (11) as:
this is the basic reproductive number for S2IR model.
2.3 The S3IR, The S4IR and S5IR Models
Here we show the S3IR, the S4IR and S5IR models where there are three, four and five groups of susceptibles, respectively. With these models we do the same process, and we only show the basic reproductive number.
this is the basic reproductive number for S3IR model.
this is the basic reproductive number for S4IR model.
and this is the basic reproductive number for S5IR model.
2.4 The SnIR Model
Theorem. For the equations system given by (1). The infection-free equilibrium is locally stable if R 0 < 1, and is unstable if R 0 > 1, where:
We can probe the theorem looking the inequalities corresponding to stability conditions for each system previously considered, we have listed in the Fig. 1. Using mechanized induction we obtain the general expression for the stability conditions for a system with (n + 2) equations.
We can represent into a schematic diagram, the deductive reasoning using “MR”.
Here we have an idea for the MR, it finds the similar components in each item and it has a viewer or a detector that find the sequential form for the dissimilar parts. It is just an idea, we believe this system have to be improved by the scientific community.
3 CA and MR Applied to the SNIMR Epidemic Model
The SnImR epidemic model is made by a group of equations which has n groups of susceptible individuals and m groups of infected people, this system is illustrated in the next equations:
p i is defined in (3). This is a system with (n + m + 1) equations and each constant were defined in the last model.
3.1 The SI2R Model
Like in all cases, we analyze a particular case with one group of susceptibles and two groups of infectives. The following equations describe this case:
Solving the system for the infection-free equilibrium, we find:
We generate a Jacobian as in the others cases. After, we substitute the infection-free equilibrium point and we find the eigenvalues for the system,
The stability conditions shall satisfy,
Moreover the inequalities in (29) can be written like the others cases as:
Here, we have the two basic reproductive numbers for SI2R model.
3.2 The S2I2R Model
We analyze a particular case where we have two groups of susceptible and two groups of infective, in addition we solve the system for the infection-free equilibrium and we find:
Also we generate a Jacobian; we substitute the infection-free equilibrium point and find the eigenvalues for this system:
In like manner that we obtained the last reproductive numbers, we obtain the two basic reproductive numbers for S2I2R model:
3.3 The SnImR Model
Theorem. For the equations system given by (25). The infection-free equilibriums are locally stable if the reproductive numbers of infection R 0,j < 1, and is unstable if R 0,j > 1, with j from 1 to m, where:
To prove the theorem, we used mechanized induction starting from the particular results previously obtained in (30) and (33). Finally, we find the general solution for the basic reproductive numbers for the SnImR model according with (Fig. 2).
4 CA and MR Applied to the Staged Progressive SIMR Epidemic Model
At this moment, we are concerned in the analysis of the staged progressive SImR epidemic model. This has m groups of infected; in this case the infection is staged and progressive, which is described by next system:
with the restriction for j is: 2 ≤ j ≤ m. And where the rate of infection is:
This is a system with (m + 2) equations and all the constants were described before.
4.1 The Staged Progressive SI2R Model
Now, we analyze a particular case with one group of susceptible and two groups of infectives. The following equations describe this case:
Solving the system for the infection-free equilibrium, we find:
We generate a Jacobian coming off the equations system and substituting the infection-free equilibrium point:
Now we obtain the characteristic polynomial:
From this polynomial, we obtain the basic reproduction number using the Routh-Hurwitz theorem:
4.2 The Staged Progressive SI3R Model
We analyze a particular case with three groups of infectives. We solve this system in the same way that we did in the Section 4.1. For this model the basic reproduction number that we find is:
4.3 Staged Progressive SImR Model
Theorem. For the equations system given by (35). The infection-free equilibrium is locally stable if the reproductive numbers of infection R 0 < 1, and is unstable if R 0 > 1, with j from 1 to m, where:
To prove the theorem, we used mechanized induction starting from the particular results previously obtained. To conclude, we find the general solution for the basic reproductive numbers for the Staged Progressive SI m R model according with (Fig. 3 ).
5 Conclusions
We finally obtain three theorems which can help us to demonstrate that CAS+MR are important tools for solving problems in every situation that mathematics could model. The theorems are useful to make strategies to fight against epidemic diseases in the future biological dangers.
Due to use CAS, in our case “Maple 11”, we can proceed to solve the mathematical problem and we can obtain results very fast that without them could take us too much time.
The use of CAS+MR can help in teaching and learning the mathematics to engineering, whose don’t have time and need to give quickly solutions. It can be implemented in engineer programs.
Through the MR we already found the general forms for the infection free equilibrium, we can see the importance of developing software it can do the mechanized reasoning automatically.
References
C.W. Brown, M.E. Kahoui, D. Novotni, A. Weber, Algorithmic methods for investigating equilibria in epidemic modeling. J. Symb. Comput. 41, 1157–1173 (2006)
J.M. Hyman, J. Li, Differential susceptibility epidemic models. J. Math. Biol. 50, 626–644 (2005)
J.A.M. Taborda, Epidemic thresholds via computer algebra, MSV 2008, pp. 178–181, http://dblp.uni-trier.de/rec/bibtex/conf/msv/Taborda08
D. Hincapié et al., Epidemic thresholds in SIR and SIIR models applying an algorithmic method, Lectures Notes Bioinformatics, 5354, 119–130, 2008
S.B. Hsua, Y.H. Hsiehb, On the role of asymptomatic infection in transmission dyanamics of infectious diseases, Bull. Math. Biol. 70, 134–155 (2008)
J.M. Hyman, J. Li, An intuitive formulation for the reproductive number for spread of diseases in heterogeneous populations. Math. Biosci. 167, 65–86 (2000)
N.T.J. Bailey, The Mathematical Theory of Epidemics, 1957
Acknowledgments
The authors would like to thank Prof. Dr. Eng. Andrés Sicard and Prof. Mario Elkin Vélez for keeping company in this work in the Logic and Computation group, Prof. Eng. Maurio Arroyave for supporting this work from the engineering physics program. They are also very grateful for Prof. Dr. Eng. Félix Londoño for his help in the World Congress on Engineering and Computer Sciences and the Student Organization and Ms. Angela Echeverri for the economical assistance in the trip to the congress. This research was support by the Logic and Computation group at EAFIT University.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer Science+Business Media B.V.
About this chapter
Cite this chapter
Cano, D.C. (2010). Generalizations in Mathematical Epidemiology. In: Ao, SI., Rieger, B., Amouzegar, M. (eds) Machine Learning and Systems Engineering. Lecture Notes in Electrical Engineering, vol 68. Springer, Dordrecht. https://doi.org/10.1007/978-90-481-9419-3_43
Download citation
DOI: https://doi.org/10.1007/978-90-481-9419-3_43
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-9418-6
Online ISBN: 978-90-481-9419-3
eBook Packages: EngineeringEngineering (R0)