Pith. sign in

REVIEW 5 major objections 4 minor 1 cited by

Determining a parabolic-elliptic-elliptic system by boundary observation of its non-negative solutions under chemotaxis background

T0 review · 5 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proves that boundary traces and final-time values of non-negative solutions of a parabolic-elliptic-elliptic chemotaxis system uniquely determine all model coefficients, including the attraction and repulsion sensitivities…

desk verdict A genuinely new inverse problem and plausible strategy, but the proof's elliptic initial data and zero-state linearization are load-bearing flaws; reject as is. read the letter →

arxiv 2509.04850 v1 pith:GTS4BGTS submitted 2025-09-05 math.AP q-bio.CBq-bio.SC

Yuhan Li , Hongyu Liu , Catharine W. K. Lo This is my paper · ORCID
classification math.APq-bio.CBq-bio.SC MSC 35R3092-1035Q9235B0935K9935J99
keywords parabolic-elliptic-ellipticsystemchemotaxisinverseproblemuniqueidentifiabilityboundarymeasurementsnon-negativesolutionshigh-ordervariationmultiplicativeseparableform
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

This paper establishes a unique identifiability result for a coupled parabolic-elliptic-elliptic chemotaxis system: for two admissible parameter configurations, if the boundary traces and final-time values of their non-negative solutions agree for every admissible initial triple $(f,g,h)$, then the two configurations coincide in $\Omega\times\mathbb{R}$. The recovered parameters include the chemotactic sensitivities $\chi$ and $\xi$, the logistic growth coefficients $r$ and $\mu$, and every Taylor coefficient of the nonlinear chemical kinetic terms $G$ and $H$. A sympathetic reader would care because this is the first inverse-problem treatment of a mixed parabolic-elliptic-elliptic biological model, and it makes the measurement map a one-to-one readout of the model's parameters. The proof proceeds by high-order linearization around a constant solution, with each coefficient isolated by a suitable choice of initial data and converted into an integral identity tested against complex geometric optics solutions. A corollary applies the general theorem to an attraction-repulsion chemotaxis model with logistic growth, recovering the full parameter set $\{\chi,\xi,r,\mu,\alpha,\beta,\gamma,\delta\}$.

What carries the argument

The load-bearing machinery is high-order variation: initial data are expanded around a known constant solution $(u_0,v_0,w_0)$ as $f=u_0+\varepsilon f_1+\frac{\varepsilon^2}{2}f_2+\cdots$, and the $\varepsilon^k$-derivatives of the solution satisfy linear parabolic-elliptic-elliptic systems. Each coefficient is recovered by choosing $f_1,g_1,h_1$ so that only the term of interest survives, then testing the resulting integral identity against special test functions. Two auxiliary tools carry the argument: Lemma 2.1, which provides parabolic test solutions of the form $u(x,t)=e^{\theta t}l(x;\theta)$ with $l$ a Neumann eigenfunction whose gradient does not vanish on any open set, and Lemma 2.2, a uniqueness result for multiplicative-separable functions $A(x)=A_1(x_1,\dots,x_{n-1})A_2(x_n)$ that turns equal Fourier-type integrals into pointwise equality of the factors. Complex geometric optics solutions of the linearized elliptic and parabolic equations supply the oscillatory kernels $e^{\zeta\cdot x}$ and $e^{(|\zeta|^2-r)t-i\zeta\cdot x}$, and the inverse Fourier transform together with the fundamental theorem of calculus converts vanishing time integrals into pointwise identities.

What would settle it

For the $\tau=0$ system (1.18), take positive initial data $(f,g,h)$ and solve the elliptic equations $0=\Delta v+\alpha u-\beta v$ and $0=\Delta w+\gamma u-\delta w$ at $t=0$; if the resulting $v(0),w(0)$ differ from the prescribed $g,h$, then the premise behind recovering $\alpha_{01}$ and $\beta_{01}$ fails. A decisive test is a numerical search over admissible configurations $A_1\neq A_2$ for which the measurement maps $M^+_{A_1}$ and $M^+_{A_2}$ coincide on all non-negative admissible initial data; one such pair would refute Theorem 1.5, while a systematic absence of such pairs would support it.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.5: under the paper's admissibility classes ($F\in\mathcal{A}$, $G\in\mathcal{B}$, $H\in\mathcal{C}$), with $\chi,\xi$ and all high-order variations independent of one spatial variable, equality of the measurement maps $M^+_{A_1}(f,g,h)=M^+_{A_2}(f,g,h)$ for all admissible non-negative initial data $(f,g,h)\in C^{2+\alpha}(\Omega)^3$ implies $A_1=A_2$ in $\Omega\times\mathbb{R}$. In words, the map from parameters to boundary and final-time observations is injective on the admissible set. The proof recovers the coefficients in layers: first-order variation gives $r$, $\alpha_{01}$, $\alpha_{10}(x)$, $\beta_{01}$, $\beta_{10}(x)$; second-order variation gives $\chi$, $\xi$, $\mu$, and the quadratic Taylor coefficients $\alpha_{11},\alpha_{20},\alpha_{02},\beta_{11},\beta_{20},\beta_{02}$; induction on the order of variation gives all higher Taylor coefficients of $G$ and $H$. The same framework, specialized to the logistic attraction-repulsion model, yields Corollary 1.6: the entire biological parameter set $\{\chi,\xi,r,\mu,\alpha,\beta,\gamma,\delta\}$ is uniquely determined by the measurement map, for both the parabolic-elliptic-elliptic ($\tau=0$) and fully parabolic ($\tau=1$) regimes.

Load-bearing premise

The proof assumes that, in the $\tau=0$ elliptic case, the initial values $g$ and $h$ of the chemical concentrations can be prescribed freely and that positivity of the linearized elliptic solutions $v^{(I)}$ and $w^{(I)}$ follows from positive $g_1$ and $h_1$; in a parabolic-elliptic-elliptic system those concentrations solve elliptic equations and are determined by $u$ and boundary conditions, so arbitrary positive initial values are generally not compatible and the positivity step is not guaranteed.

Editorial extensions

If this is right

  • If Theorem 1.5 holds, the full parameter set of the parabolic-elliptic-elliptic chemotaxis model is identifiable from boundary and final-time observations, so unknown chemotaxis and kinetic coefficients can in principle be reconstructed without interior measurements.
  • The identifiability applies simultaneously to the $\tau=0$ (mixed-type) and $\tau=1$ (fully parabolic) regimes, so the result is stable across time-scale models of the same biological process.
  • Spatially dependent Taylor coefficients are recoverable, not just constants, under the multiplicative-separability condition, extending the inverse theory beyond constant-coefficient biological models.
  • Because only non-negative solutions are used, the identifiability statement is compatible with biologically meaningful population densities and avoids requiring sign-changing inputs.
  • The recovery of $r$, $\chi$, $\xi$, $\mu$ by second-order variations, and all higher-order coefficients by induction, gives a complete parameter identifiability result rather than partial recovery of a few coefficients.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: a natural next problem is stability; the paper establishes uniqueness but not quantitative stability, so the practical value for numerical reconstruction depends on an additional Lipschitz or logarithmic stability estimate that is not derived here.
  • Editorial inference: the admissibility conditions (independence of one spatial variable and multiplicative separability of higher Taylor coefficients) are used by the Fourier-transform argument of Lemma 2.2, so whether identifiability holds for fully general $x$-dependent coefficients is an open question the current proof does not settle.
  • Editorial inference: for the $\tau=0$ case the inverse problem may need to be restated over the set of attainable initial data, since the elliptic equations for $v$ and $w$ determine their values from $u$ and boundary conditions; the proof's free choice of positive $g_1,h_1$ is the point where the mixed-type structure pushes back.
  • Editorial inference: the same high-order-variation-plus-CGO strategy looks transferable to other biological systems with elliptic components, such as chemotaxis models with nutrient or oxygen quasi-steady-state equations, provided the elliptic solution inherits the needed positivity and the initial values are compatible.
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.

Referee Report

5 major / 4 minor

Summary. The paper studies the inverse problem of determining the coefficients χ, ξ, and the Taylor coefficients of the nonlinearities F, G, H in a parabolic-elliptic-elliptic chemotaxis system from boundary traces and final-time measurements of non-negative solutions. The main result, Theorem 1.5, asserts that equality of the measurement maps M_A1^+ and M_A2^+ over all admissible initial data implies A_1 = A_2 in Ω × R. The proof is based on high-order linearization around a constant solution, construction of CGO solutions, Fourier inversion, and a multiplicative-separability uniqueness lemma, with a final application to an attraction-repulsion chemotaxis model with logistic growth in Corollary 1.6.

Significance. If Theorem 1.5 were valid, it would be a substantial contribution: unique recovery of all coefficients and all Taylor coefficients in a mixed parabolic-elliptic-elliptic system from boundary data appears to be new and would extend the authors' earlier parabolic-system results. The paper is clearly organized, states explicit admissible classes, and attempts to handle the nonlinear structure through high-order variations. However, several load-bearing technical assumptions in the τ = 0 proof are not justified, and at least one is inconsistent with the stated generality of the theorem. The central claim is therefore not established by the arguments presented.

major comments (5)
  1. [§2.2, Eqs. (2.5)-(2.7)] In the τ = 0 case, v and w satisfy elliptic equations at every time, so their initial values g and h cannot be prescribed independently of f. The first-order system (2.7) nevertheless imposes both an elliptic equation and an initial condition on v^(I) and w^(I); for a generic positive g1 no solution exists. For instance, with f1 = 0, g1 ≡ 1, and G = αu − βv, the elliptic compatibility condition at t = 0 reduces to β = 0. Consequently, the variation of g1 and h1 used to recover α01 in (2.18)-(2.20) and β01 in (2.27)-(2.29) has no admissible test functions, and the conclusions α01_1 = α01_2 and β01_1 = β01_2 are unsupported. All later steps that reuse v^(I) and w^(I) inherit this gap.
  2. [§2.2, linearization point] The proof states that '(u_j, v_j, w_j) = (0,0,0) is the solution ... when ε = 0', but Definitions 1.1-1.4 and Theorem 1.5 assert expansion around an arbitrary known non-negative constant solution (u0,v0,w0). Linearizing at (0,0,0) gives the first-order u-equation in (2.7), whereas the linearization around a general constant state would contain the terms −χ_j u0 Δv^(I) + ξ_j u0 Δw^(I) and the coefficient r_j − 2μ_j u0. These omitted terms and the incorrect coefficient are load-bearing for the recovery of r, χ, ξ, and μ in §2.2-§2.3. In particular, for the logistic nonlinearity the only constant solutions are u0 = 0 and u0 = r/μ, so the claimed generality around arbitrary constant solutions is not achieved.
  3. [§2.3, Eqs. (2.34)-(2.36)] The device 'By controlling the initial data so that Δv^(I)(x,t) = Δw^(I)(x,t) = 0' is not available for the elliptic components. The elliptic equation in (2.7) determines Δv^(I) as −α10(x)u^(I) − α01 v^(I), and v^(I) itself is not freely prescribable; imposing Δv^(I) = 0 forces v^(I) = −(α10/α01)u^(I) up to a harmonic function, which is generally incompatible with the Neumann boundary condition and with the independent positive choices of initial data used elsewhere. The subsequent derivation of (2.35)-(2.36) and the recovery of χ, ξ, and μ in (2.38)-(2.46) therefore lack a valid input-data construction.
  4. [§2.4, higher-order recovery] The proof of recovery of all higher-order coefficients is not carried out. After the second-order step, §2.4 states only that 'The main idea ... is mathematical induction, based on the k-th variation' and then declares the proof complete. No induction hypothesis, no compatibility conditions for the elliptic components at order k, and no construction of admissible test data are given. Since Theorem 1.5 asserts equality of the full Taylor coefficients in Ω × R, this is an essential missing part of the proof, not a mere presentation detail.
  5. [§2.1, Lemma 2.2] Lemma 2.2, used in §2.3 to recover α02, α11, β02, and related coefficients, is proved by separating variables as if Ω were a rectangular product Ω' × (a,b). For a general bounded Lipschitz domain the cross-section {x_n : (x',x_n) ∈ Ω} depends on x', so the factorization in (2.2) and the assertion that the x'-integrals are Fourier transforms are not valid. The lemma may hold for product domains, but as stated it is not established for the domains permitted by Theorem 1.5.
minor comments (4)
  1. [§2.4, line 1] The condition 'ℓ ∈ N, N > 2' should read 'ℓ ∈ N, ℓ ≥ 3'.
  2. [Abstract and introduction] The citation markers such as '[1111–1313]' and '[ 11,22,55,66,2727]' appear to be formatting artifacts of the bibliography; they should be cleaned to standard reference numbers.
  3. [Eq. (2.46)] The exponential factor e^{(2μ2+|ζ|^2−r)t} appears to be a typo; the exponent should involve the eigenvalue parameter θ of the heat-equation mode in (2.11), not μ2.
  4. [§1.3, Theorem 1.5] The quantifier 'For any (f,g,h) ∈ C^{2+α} × C^{2+α} × C^{2+α}' is too broad for the τ = 0 case because not every such triple is admissible for the elliptic equations; the statement should restrict to compatible initial data.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the uniqueness proof is self-contained given its assumptions; only a minor auxiliary self-citation appears in Lemma 2.1.

full rationale

The paper's central claim is an identifiability theorem: if the boundary-trace and final-time measurement maps for two parameter configurations agree on all admissible initial data, then the coefficients coincide. The proof proceeds by high-order variation around a constant solution, deriving linearized elliptic and parabolic systems, and then isolating each coefficient difference through carefully chosen initial perturbations and CGO test functions followed by Fourier inversion or the multiplicative-separable uniqueness lemma. None of the recovered coefficients is used to define the measurement map, and no fitted parameter is later renamed as a prediction. Lemma 2.1 is cited from the authors' prior paper [17], but it is a parameter-free supporting fact about spectral representations of solutions to a linear heat equation and does not contain the target identifiability conclusion. Lemma 2.2 is partly credited to [23], an independent tomography result by one of the present authors with a co-author outside this paper, and it is again a supporting uniqueness fact, not a restatement of the main theorem. The proof's apparent difficulties with freely prescribed positive initial data for the elliptic components and with positivity of the linearized elliptic solutions v^(I), w^(I) are substantive correctness concerns, but they are not circularity: they concern whether the admitted inputs are compatible with the model, not whether the conclusion has been assumed as an input. For that reason the circularity score is low, reflecting only the minor self-citation in an auxiliary lemma.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The paper introduces no fitted parameters or invented physical entities. Its proof rests on the unproved differentiability of the solution map, invalid free initial data for the elliptic components, an implicit u0=0 in the first-order linearization, and several structural admissibility assumptions that are tailored to the CGO/Fourier argument.

assumptions (7)
  • ad hoc to paper The solution map S for (1.14) is Fréchet differentiable to arbitrary order with respect to initial data at the chosen constant solution.
    Invoked in §2.2 around (2.6) to define first- and higher-order variation systems; no proof or reference is given for this differentiability for the nonlinear mixed-type system.
  • ad hoc to paper The elliptic components v and w can be assigned arbitrary non-negative initial data g,h independently, and their linearizations inherit positivity from these initial data via a maximum principle.
    Used in §2.2, equations (2.18)-(2.20) and (2.27)-(2.29); for τ=0, v and w satisfy elliptic equations with no time derivative, so their initial values are constrained by compatibility and positivity does not follow from g1>0.
  • ad hoc to paper For the τ=0 case, the constant state about which the system is linearized has u0=0, so the chemotaxis coupling terms vanish in the first-order u-equation.
    System (2.7) contains no χ or ξ terms; this is only consistent with the formal linearization if u0=0, although the paper says it expands around arbitrary constant solutions (Section 1.3, Theorem 1.5).
  • domain assumption The nonlinearities F,G,H belong to the admissible classes A,B,C in Definitions 1.1-1.4, including constancy/independence conditions on first-order Taylor coefficients and multiplicative separability of higher-order coefficients.
    These assumptions are stated in Theorem 1.5 and are needed for the Fourier/CGO recovery in Lemma 2.2; they substantially restrict the class of systems covered.
  • domain assumption Coefficients χ, ξ, and all orders of u, v, w under high-order variation are independent of one spatial variable.
    Stated in Theorem 1.5; the proof relies on this one-dimensional symmetry for the separation-of-variables argument in Lemma 2.2.
  • domain assumption All solutions achieve the regularity C^{1+α/2,2+α}_0(Q) and the problem is well-posed as cited from [9,29,31].
    Theorem 1.5 assumes existence of a solution in that class; well-posedness for the specific nonlinearities is taken from cited works.
  • standard math Neumann eigenfunctions of the Laplacian on Ω provide a complete family and can be selected as initial perturbations f1 while preserving non-negativity of the full initial data.
    Used in the recoveries of r, α10, β10 and μ (equations (2.14), (2.22), (2.31), (2.46)); completeness is standard but the compatibility with non-negativity of the perturbed data is not addressed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Determining a parabolic-elliptic-elliptic system by boundary observation of its non-negative solutions under chemotaxis background." pith.science (2026). https://pith.science/paper/GTS4BGTS

@misc{pith2026250904850,
  author       = {Pith},
  title        = {Pith review of: Determining a parabolic-elliptic-elliptic system by boundary observation of its non-negative solutions under chemotaxis background},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GTS4BGTS}},
  note         = {Machine review of arXiv:2509.04850}
}
read the original abstract

This paper addresses a profoundly challenging inverse problem that has remained largely unexplored due to its mathematical complexity: the unique identification of all unknown coefficients in a coupled nonlinear system of mixed parabolic-elliptic-elliptic type using only boundary measurements. The system models attraction-repulsion chemotaxis--an advanced mathematical biology framework for studying sophisticated cellular processes--yet despite its significant practical importance, the corresponding inverse problem has never been investigated, representing a true frontier in the field. The mixed-type nature of this system introduces significant theoretical difficulties that render conventional methodologies inadequate, demanding fundamental extensions beyond existing techniques developed for simpler, purely parabolic models. Technically, the problem presents formidable obstacles: the coupling between parabolic and elliptic components creates inherent analytical complications, while the nonlinear structure resists standard approaches. From an applied perspective, the biological relevance adds another layer of complexity, as solutions must maintain physical interpretability through non-negativity constraints. Our work provides a complete theoretical framework for this challenging problem, establishing rigorous unique identifiability results that create a one-to-one correspondence between boundary data and the model's parameters. We demonstrate the power of our general theory through a central biological application: the full parameter recovery for an attraction-repulsion chemotaxis model with logistic growth, thus opening new avenues for quantitative analysis in mathematical biology.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Unveiling Biological Models Through Turing Patterns

    math.AP 2025-09 reject novelty 5.0 of 10

    The paper proposes recovering all diffusion and chemotaxis parameters from the Fourier amplitudes of a single Turing pattern, but the uniqueness proof is incomplete and unvalidated.

Reference graph

Works this paper leans on

31 extracted references · 26 canonical work pages · cited by 1 Pith paper

  1. [1]

    Local and global solvability of some parabolic systems modelling chemo- taxis.Advances in Mathematical Sciences and Applications, 8:715–743, 1998

    Piotr Biler. Local and global solvability of some parabolic systems modelling chemo- taxis.Advances in Mathematical Sciences and Applications, 8:715–743, 1998

    work page 1998

  2. [2]

    Global solutions to some parabolic-elliptic systems of chemotaxis.Advances in Mathematical Sciences and Applications, 9:347–359, 1999

    Piotr Biler. Global solutions to some parabolic-elliptic systems of chemotaxis.Advances in Mathematical Sciences and Applications, 9:347–359, 1999

    work page 1999

  3. [3]

    Logarithmic stability in determining the time- dependent zero order coefficient in a parabolic equation from a partial Dirichlet-to- Neumann map

    Mourad Choulli and Yavar Kian. Logarithmic stability in determining the time- dependent zero order coefficient in a parabolic equation from a partial Dirichlet-to- Neumann map. Application to the determination of a nonlinear term.Journal de Math´ ematiques Pures et Appliqu´ ees, 114:235–261, 2018

    work page 2018

  4. [4]

    Inverse problems for cou- pled nonlocal nonlinear systems arising in mathematical biology.arXiv preprint arXiv:2407.15713, 2024

    Ming-Hui Ding, Hongyu Liu, and Catharine WK Lo. Inverse problems for cou- pled nonlocal nonlinear systems arising in mathematical biology.arXiv preprint arXiv:2407.15713, 2024

    arXiv 2024

  5. [5]

    Global behaviour of a reaction-diffusion system modelling chemotaxis.Mathematische Nachrichten, 195(1):77–114, 1998

    Herbert Gajewski, Klaus Zacharias, and Konrad Gr¨ oger. Global behaviour of a reaction-diffusion system modelling chemotaxis.Mathematische Nachrichten, 195(1):77–114, 1998

    work page 1998

  6. [6]

    Singularity patterns in a chemotaxis model

    Miguel A Herrero and Juan JL Vel´ azquez. Singularity patterns in a chemotaxis model. 1996

    work page 1996

  7. [7]

    From 1970 until present: the Keller-Segel model in chemotaxis and its consequences

    Dirk Horstmann. From 1970 until present: the Keller-Segel model in chemotaxis and its consequences. 2003

    1970

  8. [8]

    On uniqueness in inverse problems for semilinear parabolic equations

    Victor Isakov. On uniqueness in inverse problems for semilinear parabolic equations. Archive for Rational Mechanics and Analysis, 124(1):1–12, 1993. 22

    work page 1993

Show all 31 references
  1. [9]

    Global existence in a fully parabolic attraction-repulsion chemotaxis system with singular sensitivities and proliferation

    Zhan Jiao, Irena Jadlovsk´ a, and Tongxing Li. Global existence in a fully parabolic attraction-repulsion chemotaxis system with singular sensitivities and proliferation. Journal of Differential Equations, 411:227–267, 2024

    2024

  2. [10]

    Boundedness of the attraction–repulsion Keller–Segel system.Journal of Mathematical Analysis and Applications, 422(2):1463–1478, 2015

    Hai-Yang Jin. Boundedness of the attraction–repulsion Keller–Segel system.Journal of Mathematical Analysis and Applications, 422(2):1463–1478, 2015

    2015

  3. [11]

    Initiation of slime mold aggregation viewed as an instability.Journal of theoretical biology, 26(3):399–415, 1970

    Evelyn F Keller and Lee A Segel. Initiation of slime mold aggregation viewed as an instability.Journal of theoretical biology, 26(3):399–415, 1970

    1970

  4. [12]

    Model for chemotaxis.Journal of theoretical biology, 30(2):225–234, 1971

    Evelyn F Keller and Lee A Segel. Model for chemotaxis.Journal of theoretical biology, 30(2):225–234, 1971

    1971

  5. [13]

    Traveling bands of chemotactic bacteria: a theoretical analysis.Journal of theoretical biology, 30(2):235–248, 1971

    Evelyn F Keller and Lee A Segel. Traveling bands of chemotactic bacteria: a theoretical analysis.Journal of theoretical biology, 30(2):235–248, 1971

    1971

  6. [14]

    Partial data inverse problems and simultaneous recovery of boundary and coefficients for semilinear elliptic equations.Revista Matem´ atica Iberoamericana, 37(4):1553–1580, 2020

    Matti Lassas, Tony Liimatainen, Yi-Hsuan Lin, and Mikko Salo. Partial data inverse problems and simultaneous recovery of boundary and coefficients for semilinear elliptic equations.Revista Matem´ atica Iberoamericana, 37(4):1553–1580, 2020

    2020

  7. [15]

    Inverse problems for elliptic equations with power type nonlinearities.Journal de math´ ematiques pures et appliqu´ ees, 145:44–82, 2021

    Matti Lassas, Tony Liimatainen, Yi-Hsuan Lin, and Mikko Salo. Inverse problems for elliptic equations with power type nonlinearities.Journal de math´ ematiques pures et appliqu´ ees, 145:44–82, 2021

    2021

  8. [16]

    Roles of density-related diffusion and signal-dependent motilities in a chemotaxis–consumption system.Calculus of Variations and Partial Differential Equations, 63(8):195, 2024

    Genglin Li and Yuan Lou. Roles of density-related diffusion and signal-dependent motilities in a chemotaxis–consumption system.Calculus of Variations and Partial Differential Equations, 63(8):195, 2024

    2024

  9. [17]

    On inverse problems in predator-prey models.Journal of Differential Equations, 397:349–376, 2024

    Yuhan Li, Hongyu Liu, and Catharine WK Lo. On inverse problems in predator-prey models.Journal of Differential Equations, 397:349–376, 2024

    2024

  10. [18]

    On inverse problems in multi-population aggregation models.Journal of Differential Equations, 414:94–124, 2025

    Yuhan Li, Hongyu Liu, and Catharine WK Lo. On inverse problems in multi-population aggregation models.Journal of Differential Equations, 414:94–124, 2025

    2025

  11. [19]

    On the simultaneous recovery of environmental factors in the 3D chemotaxis-navier-stokes models.Communications on Analysis and Computation, 2(1):30–47, 2024

    Yuhan Li and Catharine WK Lo. On the simultaneous recovery of environmental factors in the 3D chemotaxis-navier-stokes models.Communications on Analysis and Computation, 2(1):30–47, 2024

    2024

  12. [20]

    Large-time behavior of an attraction– repulsion chemotaxis system.Journal of Mathematical Analysis and Applications, 426(1):105–124, 2015

    Ke Lin, Chunlai Mu, and Liangchen Wang. Large-time behavior of an attraction– repulsion chemotaxis system.Journal of Mathematical Analysis and Applications, 426(1):105–124, 2015

    2015

  13. [21]

    Global boundedness in a fully parabolic attraction– repulsion chemotaxis model.Mathematical Methods in the Applied Sciences, 38(12):2537–2546, 2015

    Dongmei Liu and Youshan Tao. Global boundedness in a fully parabolic attraction– repulsion chemotaxis model.Mathematical Methods in the Applied Sciences, 38(12):2537–2546, 2015

    2015

  14. [22]

    Determining a parabolic system by boundary observation of its non-negative solutions with biological applications.Inverse Problems, 40(2):025009, 2024

    Hongyu Liu and Catharine WK Lo. Determining a parabolic system by boundary observation of its non-negative solutions with biological applications.Inverse Problems, 40(2):025009, 2024

    2024

  15. [23]

    Determining both sound speed and internal source in thermo-and photo-acoustic tomography.Inverse Problems, 31(10):105005, 2015

    Hongyu Liu and Gunther Uhlmann. Determining both sound speed and internal source in thermo-and photo-acoustic tomography.Inverse Problems, 31(10):105005, 2015. 23

    2015

  16. [24]

    Classical solutions and steady states of an attraction– repulsion chemotaxis in one dimension.Journal of biological dynamics, 6(sup1):31–41, 2012

    Jia Liu and Zhi-An Wang. Classical solutions and steady states of an attraction– repulsion chemotaxis in one dimension.Journal of biological dynamics, 6(sup1):31–41, 2012

    2012

  17. [25]

    Mathematical analysis of the dynamics of some reaction-diffusion models for infectious diseases.Journal of Differential Equations, 370:424–469, 2023

    Yuan Lou and Rachidi B Salako. Mathematical analysis of the dynamics of some reaction-diffusion models for infectious diseases.Journal of Differential Equations, 370:424–469, 2023

    2023

  18. [26]

    Chemotactic signaling, microglia, and Alzheimer’s disease senile plaques: Is there a connection?Bulletin of mathematical biology, 65(4):693–730, 2003

    Magdalena Luca, Alexandra Chavez-Ross, Leah Edelstein-Keshet, and Alex Mogilner. Chemotactic signaling, microglia, and Alzheimer’s disease senile plaques: Is there a connection?Bulletin of mathematical biology, 65(4):693–730, 2003

    2003

  19. [27]

    Global existence and blow-up of radial solutions to a parabolic-elliptic system of chemotaxis.Adv

    Toshitaka Nagai. Global existence and blow-up of radial solutions to a parabolic-elliptic system of chemotaxis.Adv. Math. Sci. Appl., 8:145, 1998

    1998

  20. [28]

    Volume-filling and quorum-sensing in models for chemosensitive movement.Can

    Kevin J Painter, Thomas Hillen, et al. Volume-filling and quorum-sensing in models for chemosensitive movement.Can. Appl. Math. Quart, 10(4):501–543, 2002

    2002

  21. [29]

    Well-posedness for a model derived from an attraction– repulsion chemotaxis system.Journal of Mathematical Analysis and Applications, 423(1):497–520, 2015

    Renkun Shi and Weike Wang. Well-posedness for a model derived from an attraction– repulsion chemotaxis system.Journal of Mathematical Analysis and Applications, 423(1):497–520, 2015

    2015

  22. [30]

    Competing effects of attraction vs

    Youshan Tao and Zhi-An Wang. Competing effects of attraction vs. repulsion in chemo- taxis.Mathematical Models and Methods in Applied Sciences, 23(01):1–36, 2013

    2013

  23. [31]

    A parabolic–elliptic–elliptic attraction– repulsion chemotaxis system with logistic source.Journal of Mathematical Analysis and Applications, 455(1):650–679, 2017

    Jie Zhao, Chunlai Mu, Deqin Zhou, and Ke Lin. A parabolic–elliptic–elliptic attraction– repulsion chemotaxis system with logistic source.Journal of Mathematical Analysis and Applications, 455(1):650–679, 2017. 24

    2017

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.