Pith. sign in

REVIEW 45 references

Continuous and discrete compartmental models for infectious disease

T0 review · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read A curve-fitting comparison of cellular automaton and ODE versions of SI, SIR, and SEIR models finds power-law versus exponential early growth and proposes a hyperbolic tangent fit for both.

arxiv 2504.13953 v1 pith:T2Y5QWTB submitted 2025-04-16 q-bio.PE physics.bio-ph

classification q-bio.PEphysics.bio-ph
keywords modelsmathematicaldiseasefunctiongrowthdiseasesexponentialhyperbolic
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Epidemiologists often describe an outbreak with compartmental models that divide people into susceptible, infected, and recovered groups. This paper compares two ways of running those models: a cellular automaton on a grid, where each individual is a cell that infects nearby neighbors with some probability, and a set of differential equations, where the whole population is treated as one well-mixed fluid. The authors simulate SI, SIR, and SEIR versions with both approaches and look at how the number of infected people grows at the beginning of the outbreak. In the grid model, the early increase looks like a power law (a line on a log-log plot), while in the differential-equation version it looks exponential (a line on a semi-log plot), at least for two of the three models. They then fit a hyperbolic tangent function to every early outbreak curve and obtain high R-squared values, concluding that this sigmoid shape can represent both growth styles.

The main issue is that the comparison is descriptive curve fitting, not a derivation. The parameters in the fit are chosen after the data are generated, and no alternative functions are tested. The paper also sets the infection probability equal in both representations, but in the grid model this probability acts per infected neighbor per time step while in the equations it acts as a per-capita contact rate; equating the numbers does not make the models dynamically equivalent. For the SEIR differential-equation model the exponential fit is actually poor (R-squared 0.75), which contradicts the abstract's claim that ODE growth is well represented by an exponential. The hyperbolic tangent result is essentially a tautology: any smooth S-shaped curve can be fitted by a bounded sigmoid with enough parameters.

Extended reading notes

Core claim

The abstract states: 'the CA yields a power-law growth, while the ODE growth rate is well-represented by an exponential function' and 'a substantial contribution of our work is using a hyperbolic tangent to fit the initial growth of infected individuals for all the considered models.' If correct, the paper would show that a single sigmoid function describes epidemic onset in both discrete spatial and continuous mean-field frameworks.

Load-bearing premise

The assumption that using numerically equal parameter values in both representations makes the models comparable: in the CA, beta acts as a per-neighbor infection probability per time step, while in the ODE it is a per-capita contact rate per unit time (Tables 1-3 and the text 'once selected the parameters values, they need to be equal for both approaches'). If the rates are not truly equivalent, the observed difference between power-law and exponential growth could reflect a mismatch in transmission speed rather than an intrinsic property of discrete versus continuous representation.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Assumptions & free parameters 8 free parameters · 5 assumptions · 0 invented entities

The paper's central claims rest on arbitrary parameter choices, in-sample curve fits, and an unexamined equivalence between CA and ODE rate parameters. No new entities are introduced.

free parameters (8)
  • Beta (SI, SIR, SEIR) = 0.1, 0.2, 0.25
    Chosen arbitrarily; set equal for CA and ODE to allow comparison, but the meaning differs between per-neighbor probability in CA and per-capita contact rate in ODE.
  • Gamma (SIR, SEIR) = 0.1
    Recovery probability/rate, chosen arbitrarily.
  • Omega (SEIR) = 0.2
    Incubation rate, chosen arbitrarily.
  • Initial infected/exposed fraction = 0.01 (100 cells in CA; 0.01 in ODE)
    Selected to match between representations, but the spatial arrangement in CA is random.
  • Fitting window cutoff = I = 0.15 N
    All fits are limited to the early phase up to 15 percent of the population; the cutoff is chosen by hand.
  • Power-law K and B for CA fits = SI: K=0.007, B=1.09; SIR: K=0.017, B=0.87; SEIR: K=0.0069, B=1.21
    Fitted to the simulated CA i(t) curves in Sections 3, 4, and 5.
  • Exponential K and B for ODE fits = SI: K=0.01076, B=0.09262; SIR: K=0.0175, B=0.0597; SEIR: K=0.00077, B=0.116
    Fitted to the simulated ODE i(t) curves in Sections 3, 4, and 5.
  • Tanh parameters alpha, xi, phi = CA SI: alpha=0.115, xi=0.108, phi=-1.27; ODE SI: alpha=0.500, xi=0.050, phi=-2.298; CA SIR: alpha=0.090, xi=0.150…
    Three parameters fitted to each early i(t) curve; these are the basis for the claimed hyperbolic tangent result.
assumptions (5)
  • standard math Standard SIR and SEIR ODE compartmental equations and their analytic solutions
    Used in Eqs. (1)-(4), (8)-(10), and (11)-(14).
  • domain assumption Von Neumann neighborhood and periodic boundary conditions in CA
    Defines the local interaction structure of the cellular automaton.
  • domain assumption Homogeneous mixing assumption in ODE models
    Standard mean-field assumption in compartmental ODEs.
  • ad hoc to paper Equivalence of numerically equal beta, gamma, and omega across CA and ODE
    The paper sets parameters equal across representations but does not justify that the per-neighbor probability in CA matches the per-capita rate in ODE; the comparison of growth rates relies on this.
  • standard math Series expansion and exponential representation of the hyperbolic tangent
    Used in the Conclusions to argue tanh captures both power-law and exponential growth.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Continuous and discrete compartmental models for infectious disease." pith.science (2026). https://pith.science/paper/T2Y5QWTB

@misc{pith2026250413953,
  author       = {Pith},
  title        = {Pith review of: Continuous and discrete compartmental models for infectious disease},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T2Y5QWTB}},
  note         = {Machine review of arXiv:2504.13953}
}
read the original abstract

The study of infectious disease propagation is essential for understanding and controlling epidemics. One of the most useful tools for gaining insights into the spread of infectious diseases is mathematical modelling. In terms of mathematical epidemiology, the main models are based on compartments, such as SI, SIR, and SEIR. These models offer mathematical frameworks for representing the proliferation dynamics of various diseases, for instance flu and smallpox. In this work, we explore these models using two distinct mathematical approaches, Cellular Automata (CA) and ODEs. They are able to reproduce the spread dynamics of diseases with their own individuality. CA models incorporate the local interaction among individuals with discrete time and space, while ODEs provide a continuous and simplified view of a disease propagation in large and homogeneous populations. By comparing these two approaches, we find that the shape of the curves of all models is similar for both representations. Although, the growth rates differ between CA and ODE. One of our results is to show that the CA yields a power-law growth, while the ODE growth rate is well-represented by an exponential function. Furthermore, a substantial contribution of our work is using a hyperbolic tangent to fit the initial growth of infected individuals for all the considered models. Our results display a strong correlation between simulated data and adjusted function. We mainly address this successful result by the fact that the hyperbolic function captures both growing: the power-law (when considered the first terms of infinite sums) and combinations of exponential (when the hyperbolic function is written via exponential). Therefore, our work shows that when modelling a disease the choice of mathematical representation is crucial, in particular to model the onset of an epidemic.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 45 canonical work pages

  1. [1]

    Princeton University Press, Princeton (2008)

    Keeling, M.J., Rohani, P.: Modeling Infectious Diseases in Humans and Animals, 1st edn. Princeton University Press, Princeton (2008)

  2. [2]

    Oxford University Press, Oxford (1991)

    Anderson, R.M., May, R.M.: Infectious Diseases of Humans: Dynam ics and Control. Oxford University Press, Oxford (1991)

  3. [3]

    Bjornstad, O.N.: Epidemics: Models and Data Using R, 1 ª ed., Springer Nature Switzerland AG, Cham, Switzerland (2018)

  4. [4]

    Plos Medicine 6, 1000139 (2009)

    Cummings, D.A.T., Iamsirithaworn, S., Lessler, J.T., McDermott, A., Prasan- thong, R., Nisalak, A., Jarman, R.G., Burke, D.S., Gibbons, R.V.: The impa ct of the demographic transition on dengue in thailand: Insights from a statistical analysis and mathematical modeling. Plos Medicine 6, 1000139 (2009)

  5. [5]

    Chaos 30, 041102 (2020)

    Manchein, C., Brugnago, E.L., Silva, R.M., Mendes, C.F.O., Beims, M.W.: S trong correlations between power-law growth of covid-19 in four contine nts and the inefficiency of soft quarantine strategies. Chaos 30, 041102 (2020)

  6. [6]

    International Journal of Biomathematics 11, 1850108 (2018)

    Yamazaki, K.: Threshold dynamics of reaction–diffusion partial diff erential equations model of ebola virus disease. International Journal of Biomathematics 11, 1850108 (2018)

  7. [7]

    Applied M athematics and Computation 186, 193–202 (2007)

    White, S.H., Rey, A.M., S´ anchez, G.R.: Threshold dynamics of react ion–diffusion partial differential equations model of ebola virus disease. Applied M athematics and Computation 186, 193–202 (2007)

  8. [8]

    Chaos, Solitons and Fractals 139, 110058 (2020)

    Wang, P., Zheng, X., Li, J., Zhu, B.: Prediction of epidemic trends in c ovid-19 with logistic model and machine learning technics. Chaos, Solitons and Fractals 139, 110058 (2020)

Show all 45 references
  1. [9]

    Nature Medicine 26, 1417–1421 (2020)

    Hoertel, N., Blachier, M., Blanco, C., Olfson, M., Massetti, M., Rico, M .S., Limosin, F., Leleu, H.: A stochastic agent-based model of the sars- cov-2 epidemic in france. Nature Medicine 26, 1417–1421 (2020)

  2. [10]

    Reviews of Modern Physics 87, 925 (2015)

    Pastor-Satorras, R., Castellano, C., Mieghem, P., Vespignani, A .: Epidemic processes in complex networks. Reviews of Modern Physics 87, 925 (2015)

  3. [11]

    Mat hematical Biosciences 124, 83–105 (1994)

    Allen, L.J.S.: Some discrete-time si, sir, and sis epidemic models. Mat hematical Biosciences 124, 83–105 (1994). 15

  4. [12]

    Mathematical Biosciences 163, 1–33 (2000)

    Allen, L.J.S., Burgin, A.M.: Comparison of deterministic and stochas tic sis and sir models in discrete time. Mathematical Biosciences 163, 1–33 (2000)

  5. [13]

    The Lancet 167, 227–230 (1906)

    Hamer, W.: The treatment of typhoid fever. The Lancet 167, 227–230 (1906)

  6. [14]

    Proceedings of the royal society of London

    Kermack, W.O., McEndrick, A.G.: A contribution to the mathematic al theory of epidemics. Proceedings of the royal society of London. Series A 115, 700–721 (1927)

  7. [15]

    Revista Bras ileira de Ensino de F ´ ısica43, 20210171 (2021)

    Batista, A.M., Souza, S.L.T., Iarosz, K.C., Almeida, A.C.L., Jr., J.D.S., G abrick, E.C., Mugnaine, M., Santos, G.L., Caldas, I.L.: Simulation of deterministic com- partmental models for infectious diseases dynamics. Revista Bras ileira de Ensino de F ´ ısica43, 20210171 (2021)

  8. [16]

    Mathematical Biosciences 125, 155–164 (1995)

    Li, M.Y., Muldowney, J.S.: Global stability for the seir model in epidem iology. Mathematical Biosciences 125, 155–164 (1995)

  9. [17]

    Journal of Computational and Applied Math ematics 229, 313–323 (2009)

    Cai, L., Li, X., Ghosh, M., Guo, B.: Stability analysis of an hiv/aids epid emic model with treatment. Journal of Computational and Applied Math ematics 229, 313–323 (2009)

  10. [18]

    Ecological Monographs 72, 169–184 (2002)

    Bjornstad, O.N., Finkenst¨ adt, B.F., Grenfell, B.T.: Dynamics of m easles epi- demics: estimating scaling of transmission rates using a time series sir model. Ecological Monographs 72, 169–184 (2002)

  11. [19]

    Journal of Economic Development, Environmen t and People 11, 5-30 (2022)

    Mohajan, D., Mohajan, H.K.: Mathematical analysis of seir model to prevent covid-19 pandemic. Journal of Economic Development, Environmen t and People 11, 5-30 (2022)

  12. [20]

    Theory of Self-reproducing Automata, Edited by Arthur W

    Neumann, J. Theory of Self-reproducing Automata, Edited by Arthur W. Burks edn., University of Illinois Press, Illinois (1966)

  13. [21]

    A New Kind of Science

    Wolfram, S. A New Kind of Science. 1st edn., Wolfram Media, Champ aign, Illinois (2002)

  14. [22]

    American Mathematical Society (2020)

    Gardner, M.: Wheels, Life and Other Mathematical Amusements , 10 edn. American Mathematical Society (2020)

  15. [23]

    Ecological Modelling 133, 209–223 (2000)

    Sirakoulis, G.C., Karafyllidis, I., Thanailakis, A.: A cellular automaton model for the effects of population movement and vaccination on epidemic prop agationa. Ecological Modelling 133, 209–223 (2000)

  16. [24]

    Applied Mathematics and Computation 40, 41–54 (1990)

    Yakowitz, S., Gani, J., Hayes, R.: Cellular automaton modeling of ep idemics. Applied Mathematics and Computation 40, 41–54 (1990)

  17. [25]

    Physica A 267, 471–486 (1999)

    Fuentes, M.A., Kuperman, M.N.: Cellular automata and epidemiologic al models with spatial dependence. Physica A 267, 471–486 (1999). 16

  18. [26]

    Physical Review Letters 87, 168102 (2001)

    Santos, R.M.Z., Coutinho, S.: Dynamics of hiv infection: A cellular au tomata approach. Physical Review Letters 87, 168102 (2001)

  19. [27]

    Physica A 499, 75–87 (2018)

    Pereira, F.M.M., Schimit, P.H.T.: Dengue fever spreading based on p robabilistic cellular automata with two lattices. Physica A 499, 75–87 (2018)

  20. [28]

    Journal of Theoretical Biology 232, 223–234 (2005)

    Beauchemin, C., Samuel, J., Tuszynski, J.: A simple cellular automat on model for influenza a viral infections. Journal of Theoretical Biology 232, 223–234 (2005)

  21. [29]

    Chaos, Solitons and Fractals 155, 111784 (2022)

    Mugnaine, M., Gabrick, E.C., Protachevicz, P.R., Iarosz, K.C., Sou za, S.L.T., Almeida, A.C.L., Batista, A.M., Caldas, I.L., Jr, J.D.S., Viana, R.L.: Control attenuation and temporary immunity in a cellular automata seir epidem ic model. Chaos, Solitons and Fractals 155, 111784 (2022)

  22. [30]

    Boyce, W.E., DiPrima, R.C.: Elementary Differential Equations and B oundary Value Problems, 8 ª ed., John Wiley and Sons, New York (2004)

  23. [31]

    Numerical Methods for Ordinary Differential Equ ations

    Butcher, J.C. Numerical Methods for Ordinary Differential Equ ations. 1st edn., John Wiley and Sons, Chichester, England (2008)

  24. [32]

    Mathematical Models in Epidemiology

    Brauer, F., Castillo-Chavez, C., Feng, Z. Mathematical Models in Epidemiology. 1st edn., Springer, New York (2019)

  25. [33]

    Scientific Repo rts 11, 16312 (2021)

    Beira, M.J., ao, P.J.S.: A differential equations model-fitting analys is of covid- 19 epidemiological data to explain multi-wave dynamics. Scientific Repo rts 11, 16312 (2021)

  26. [34]

    Cellular Automata: A Discrete Universe

    Ilachinski, A. Cellular Automata: A Discrete Universe. 1st edn., W orld Scientific Publishing Company, Singapore (2001)

  27. [35]

    Physica A 597, 127258 (2022)

    Gabrick, E.C., Protachevicz, P.R., Batista, A.M., Iarosz, K.C., Sou za, S.L.T., Almeida, A.C.L., Jr, J.D.S., Mugnaine, M., Caldas, I.L.: Effect of two vaccin e doses in the seir epidemic model using a stochastic cellular automaton . Physica A 597, 127258 (2022)

  28. [36]

    Journal of Physics : Complexity 5, 025010 (2024)

    Souza, D.L.M., Borges, F.S., Gabrick, E.C., Bentivoglio, L.E., Protac hevicz, P.R., Santos, V., Viana, R.L., Caldas, I.L., Iarosz, K.C., Batista, A.M., Kurth s, J.: Spi- ral wave dynamics in a neuronal network model. Journal of Physics : Complexity 5, 025010 (2024)

  29. [37]

    Contemporary Physics 44, 401–416 (2003)

    Rosenblum, M., Pikovsky, A.: Synchronization: From pendulum clo cks to chaotic lasers and chemical oscillators. Contemporary Physics 44, 401–416 (2003)

  30. [38]

    The Journal of Chemical Physic s 119, 6388–6395 (2003)

    Kikuchi, N., Pooley, C.M., Ryder, J.F., Yeomans, J.M.: Transport co efficients of a mesoscopic fluid dynamics model. The Journal of Chemical Physic s 119, 6388–6395 (2003). 17

  31. [39]

    SIAM R eview 42, 599–653 (2000)

    Hethcote, H.W.: The mathematics of infectious diseases. SIAM R eview 42, 599–653 (2000)

  32. [40]

    Diekmann, O., Heesterbeek, H., Britton, T.: Mathematical Tools for Understand- ing Infectious Disease Dynamics, 1st edn., p. 520. Princeton Univer sity Press (2013)

  33. [41]

    Ma thematical Biosciences 232, 31–41 (2011)

    Tian, J.P., Wang, J.: Global stability for cholera epidemic models. Ma thematical Biosciences 232, 31–41 (2011)

  34. [42]

    Journal of Mathematical Analysis and Applications 325, 36–53 (2007)

    Dalal, D., Greenhalgh, D., Mao, X.: A stochastic model of aids and c ondom use. Journal of Mathematical Analysis and Applications 325, 36–53 (2007)

  35. [43]

    Plos Neglected Tr opical Diseases 17, 0011087 (2023)

    Ospina-Aguirre, C., D.S.-P., Olivar-Tost, G., Galindo-Gonz´ ales, C., J.G.-G., Oso- rio, C.: A stochastic model of aids and condom use. Plos Neglected Tr opical Diseases 17, 0011087 (2023)

  36. [44]

    American Journal of Mathematics and Statistics 4, 231–239 (2014)

    Kuddus, A., Rahman, A., Talukder, M.R., Hoque, A.: A modified sir mo del to study on physical behaviour among smallpox infective population in ba ngladesh. American Journal of Mathematics and Statistics 4, 231–239 (2014)

  37. [45]

    Applied Mathematics 4, 84 (2013)

    Misici, L., Santarelli, F.: Epidemic propagation: An automaton mode l as the continuous sir model. Applied Mathematics 4, 84 (2013). 18

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.