Pith. sign in

REVIEW 4 major objections 4 minor 29 references

Closed-loop robust control of long-term diabetes progression via physical activity management

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read A model predictive controller that recommends physical activity can halt and reverse simulated type-2 diabetes progression, cutting total exercise effort by roughly 58% and keeping 75% of perturbed virtual patients normoglycemic.

desk verdict A clean, clearly bounded proof-of-concept for exercise-as-control in T2D, but the headline robustness number depends on tuning choices and the exercise model is unvalidated. read the letter →

arxiv 2501.12892 v1 pith:V3O5FXCX submitted 2025-01-22 eess.SY cs.SY

Pierluigi Francesco De Paola , Alessandro Borri , Fabrizio Dabbene , Pasquale Palumbo , Alessia Paglialonga This is my paper · ORCID
classification eess.SYcs.SY MSC 92C5093B45
keywords type2diabetesphysicalactivitymodelpredictivecontrolprogressioninterleukin-6insulinsensitivityrobustnessMonteCarlosimulation
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 proposes treating physical activity as a control input for type 2 diabetes: a model predictive controller recommends an exercise dose, updated over time, that is meant to stop and reverse the simulated progression of the disease. The authors modify a compact diabetes progression model so that exercise acts through an accumulated variable representing the long-term effect of interleukin-6, and they test the resulting feedback law on a nominal virtual patient and on 100 perturbed simulations. If the exercise-response model is right, the controller restores normoglycemia in the nominal case and in 75% of the perturbed cases, while using about 58% less total exercise effort than the minimum constant exercise program that prevents progression. The practical stake is a quantitative, model-based way to turn broad clinical advice into specific, time-varying session durations, and eventually to assess medical guidelines in silico.

What carries the argument

The load-bearing object is the equivalent control input $u_{\mathrm{eq}} = \bar{u}\delta/T$, the average exercise intensity over one training period, which encodes an entire exercise program (intensity, duration, frequency) as a single continuous control variable. The MPC minimizes an integral cost of glycemia and control effort, $\int (G^2+\lambda u_{\mathrm{eq}}^2)\,ds$, over a 20-period prediction horizon with a one-period control horizon, and the optimizer's output is converted into concrete session durations through the inverse map $\delta = u_{\mathrm{eq}}T/\bar{u}$. The controlled model is the Topp glucose-insulin-$\beta$-cell model augmented with a state $V_l$ for the cumulative IL-6 effect, Hill functions $\psi_1(V_l)$ and $\psi_2(V_l)$ that modulate $\beta$-cell proliferation and apoptosis, and a Michaelis-Menten term that raises insulin sensitivity during training. These modifications are what make exercise a meaningful long-term control channel in the simulations.

What would settle it

A concrete test would be to run the same MPC-generated exercise schedule on the full higher-dimensional IL-6 model from which the compact model was reduced, or on a prospective prediabetes cohort following the recommended front-loaded program of roughly 22-minute moderate sessions every two days, peaking near 250 minutes per week early on; if the full model or the cohort does not bend fasting glucose, HbA1c, or beta-cell markers back toward normoglycemia within a year, the compact-model claim of 75% success and 58% effort saving fails to transfer.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a receding-horizon model predictive control law acting on the equivalent exercise input $u_{\mathrm{eq}}$ — the time-averaged exercise intensity over the training period — can drive the modified Topp model back to the normoglycemic steady state at $G \approx 100$ mg/dl, where open-loop evolution and the constant feedforward dose $u_{\mathrm{eq}}=1$ end at the hyperglycemic steady state $G=600$ mg/dl. The MPC schedule is front-loaded: it recommends higher exercise early in the disease course and lets the dose decay toward zero as glycemia normalizes. The minimum constant input that prevents progression is $u_{\mathrm{ff,min}}=1.1$, with total control effort $\eta=401.5$; the MPC uses $\eta=168.1$, a reduction of roughly 58%. Under Monte Carlo perturbations of $\pm5\%$ on initial conditions and parameters, 75 of 100 simulated patients remain normoglycemic over the one-year horizon, while 25 progress irreversibly. The paper presents this as an encouraging proof of concept to be validated on a higher-dimensional model before clinical translation.

Load-bearing premise

The load-bearing premise is that the exercise-response equations — the Hill functions preserving beta-cell mass and the Michaelis-Menten term raising insulin sensitivity, with parameters inherited from the authors' earlier IL-6 models — correctly describe how real exercise affects human diabetes progression; if that physiological map is wrong or overstates the benefit, the controller's recommendations, the 75% success rate, and the claimed agreement with WHO advice do not transfer to patients.

Editorial extensions

If this is right

  • In the nominal simulation, MPC restores and holds glucose near $G=100$ mg/dl over the year, whereas open-loop evolution and the constant dose $u_{\mathrm{eq}}=1$ reach the hyperglycemic steady state $G=600$ mg/dl.
  • The MPC schedule's total control effort ($\eta=168.1$) is about 58% lower than the minimum constant feedforward dose ($\eta=401.5$), meaning the feedback approach achieves the same goal with substantially less cumulative exercise.
  • Converting the control signal into practice gives roughly 22 minutes of moderate-intensity exercise every two days on average, rising to weekly peaks near 250 minutes early in the course — numbers the paper reads as consistent with WHO advice to exceed 150 minutes per week for diabetes risk reduction.
  • In 100 Monte Carlo simulations with $\pm5\%$ perturbations of initial conditions and parameters, 75% of virtual patients stay normoglycemic over one year; the remaining 25% progress irreversibly, which the paper attributes to inter-individual variability in the model.

Reading between the lines

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

  • The front-loaded dose profile suggests a testable hypothesis the paper does not compare directly: a short intensive exercise induction phase followed by a low maintenance dose may dominate a constant prescription in both efficacy and adherence.
  • Because $u_{\mathrm{eq}}$ abstracts any periodic exercise program into one scalar, the same MPC architecture could be adapted to other lifestyle mediators (e.g., diet), replacing the IL-6/$V_l$ pathway with a homologous accumulation variable.
  • The 25% failure rate under only $\pm5\%$ perturbations implies that a safety-oriented extension — raising the recommended dose when simulated beta-cell mass or insulin sensitivity drops below a threshold — is a natural next layer for clinical use.
  • The claimed WHO alignment is a model-output comparison, not a clinical outcome; the paper itself identifies validation on a higher-dimensional model as the missing step before any guideline numbers should be read as patient advice.
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

4 major / 4 minor

Summary. The paper presents a model predictive control (MPC) strategy that recommends physical activity, represented by an equivalent continuous input u_eq, to prevent or reverse type-2 diabetes progression in a compact model derived by modifying the Topp model. The model adds a state V_l for the cumulative effect of IL-6 released during exercise, which modulates beta-cell proliferation/apoptosis and insulin sensitivity. The MPC law minimizes a quadratic cost with weight λ=60 over a one-year horizon and is compared with a constant feedforward input; in nominal simulations MPC restores normoglycemia while a constant u_eq=1.1 is the minimum constant input that does so, and the cumulative effort is η=168.1 versus 401.5. Monte Carlo simulations with ±5% perturbations in initial conditions and parameters give 75% success. The authors frame the work as a proof of concept toward quantitative support for exercise recommendations and explicitly acknowledge the lack of validation on high-dimensional models.

Significance. The paper addresses a gap in the control literature by treating physical activity as a control input for long-term diabetes, and the idea of an equivalent continuous input for exercise is useful. The MPC formulation on a compact model is interesting and the nominal simulations demonstrate that feedback can reduce cumulative exercise effort relative to a constant minimal-dose benchmark. However, the quantitative results are entirely in silico and depend on a model and parameters that are not validated against independent clinical data; the values λ, η, and 75% are properties of the chosen cost weight and perturbation range. The paper is transparent about its preliminary nature, which is a strength, but it should not be read as a quantitative assessment of medical guidelines in its current form.

major comments (4)
  1. [Section 2, Eq. (1e) and Table 1] The system model mixes time units. Equations (1a)-(1d) use per-day rates (R0=864 mg/dl/d, Eg0=1.44 1/d, k=432 1/d, c=0.05 1/d), but Eq. (1e) and Table 1 give SR, K_IL6, and k_s in per-minute units. With t in days, the left side of (1e), dV_l/dt, has units (pg/ml)·min/day while the right side SR/KIL6·u - k_s V_l has units pg/ml; a factor of 1440 is missing. If instead the intended time base for (1e) is minutes, the remaining equations would need rescaling, so the ODE system as written is not dimensionally consistent. This affects the time constant of V_l, the effective values of k_n,si and ζ_si, and therefore all simulations in Section 4. Please correct the units or state a consistent time base and re-run the reported results.
  2. [Section 2, Eq. (5)] The reduction of the impulsive exercise program (the pulse-train u(t) in Eq. (4)) to the constant equivalent input u_eq is asserted without proof. The text states that the fast dynamics of the full model [13,14] are at their quasi-stationary values, but no derivation or numerical comparison between the full model and the reduced model (1) is provided. Since the MPC law and the inverse map (8) both rely on this equivalence, it is load-bearing. Please provide a formal quasi-steady-state argument or a simulation-based validation that the reduced model reproduces the long-term dynamics of the full model for the exercise programs considered.
  3. [Section 4.2] The robustness claim rests on a single Monte Carlo experiment with φ=5% and λ=60, and the perturbations are drawn from the same model used to design the controller. Because the Topp model is bistable, the 75% success rate and the effort comparison η=168.1 vs 401.5 are properties of these specific choices; no sensitivity to φ or λ is reported, no confidence interval for the success rate is given, and model-form uncertainty is not addressed. Please provide a sweep over φ and λ, quantify the Monte Carlo uncertainty (e.g., a binomial confidence interval for 75/100), include a model-form sensitivity analysis or soften the robustness claim accordingly.
  4. [Section 4.1] The feedforward benchmark is restricted to the minimal constant control u_ff,min=1.1. A time-varying open-loop schedule, for example the full-horizon optimal solution of the same cost functional, could achieve similar or better glycemia with less cumulative effort than the constant benchmark. Therefore the comparison η=168.1 vs 401.5 does not by itself demonstrate the advantage of feedback. Please add a comparison with an optimal open-loop time-varying schedule, or rephrase the conclusion to 'MPC is less effort than the minimal constant feedforward'.
minor comments (4)
  1. [Section 4.2] The 100 Monte Carlo runs yield a success proportion of 75/100; reporting the 95% binomial confidence interval (about ±8.5%) would make the robustness statement more precise and would support the 'reliable robustness' phrasing.
  2. [Section 4.1, Fig. 3] The weekly exercise duration (maximum about 250 min/week) is computed with fixed intensity χ̄=60% and period T=2 days; the sensitivity of the recommendations to these assumed program parameters is not discussed.
  3. [Section 3, Eq. (7)] The cost terms x1^2 and λ u_eq^2 have different physical units, so the chosen λ=60 is meaningful only relative to the chosen units; stating a normalization basis (e.g., reference glucose and reference exercise values) would improve reproducibility.
  4. [Throughout] The paper refers to Eq. (5) as an equivalence, but it is a definition of u_eq; the quasi-stationary assumption is introduced only in the text after it. Please reword to make the modeling assumption explicit.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the MPC results are same-model simulations with explicitly disclosed tuning parameters, not predictions that reduce to their inputs by construction.

full rationale

The paper's derivation chain is: (i) adopt the Topp compact diabetes model [18]; (ii) modify it with Hill and Michaelis-Menten exercise terms inherited from the authors' prior models [13,14]; (iii) introduce the equivalent constant input ueq as the time-average of the exercise intensity; (iv) formulate an MPC problem minimizing G^2 + lambda*u^2 over a horizon N=20; and (v) simulate the closed loop under nominal and perturbed conditions, comparing cumulative effort with a constant feedforward input. No step defines its output in terms of itself. The MPC trajectory, the 75% Monte Carlo success rate, and the effort comparison are computed from the same model used to design the controller, but this is ordinary in silico model-based control evaluation rather than a disguised identity: the result depends nontrivially on the nonlinear model, the horizon, and the explicit choices lambda=60 and phi=5%, all of which the paper discloses ('parameter lambda is set equal to 60, calibrated by simulation'; 'relative variations with respect to the nominal values uniformly distributed over the interval [-phi,phi], with phi=5%'). These are openly declared tuning and scenario parameters, not fitted quantities renamed as predictions. The self-citations [13,14,15] supply the exercise-response model and the control-oriented reduction, but they do not smuggle in the conclusion: the paper explicitly frames the work as a proof of concept ('This work shows a proof of concept') and calls for validation on a high-dimensional model, acknowledging that it inherits limitations from [18]. Whether the exercise-effect model is clinically accurate is a correctness and validation risk, not a circularity, and the paper does not overclaim external clinical validation. Consequently, no circular step can be exhibited with the required specificity, and the appropriate finding is no significant circularity.

Assumptions & free parameters 5 free parameters · 4 assumptions · 1 invented entities

The central claim rests on the authors' own exercise model and on tuning choices; external benchmarks are absent.

free parameters (5)
  • lambda (cost weight) = 60
    Calibrated by simulation in Section 3; directly determines the balance between glycemia and exercise effort, hence shapes all recommended durations.
  • u_ff,min (feedforward benchmark) = 1.1
    Minimum constant equivalent input that prevents progression in the nominal model, computed in Section 4.1; used as the baseline for the control effort comparison.
  • Prediction horizon N = 20
    Chosen to match the time constant of insulin sensitivity decay (Section 3); affects the closed-loop behavior.
  • Perturbation range phi = 5%
    Uniform relative variation in initial conditions and parameters in Monte Carlo simulations (Section 4.2); the 75% success rate depends on this range.
  • Exercise-effect parameters (zeta_p, kp, zeta_a, ka, SR, K_IL6, ks, zeta_si, kn_si) = Table 1 values
    Inherited from Refs. [13,14] with no fitting or validation in this paper; they control how exercise affects beta-cell and insulin sensitivity dynamics, and thus the entire outcome.
assumptions (4)
  • domain assumption The five-state model (1) with Table 1 parameters adequately describes long-term glucose-insulin-beta-cell dynamics relevant for T2D prevention.
    Used to simulate disease course and control effects; no clinical validation on the model with exercise modifications is provided.
  • domain assumption The equivalent input u_eq = bar_u*delta/T represents the net effect of intermittent exercise sessions on the slow progression dynamics (Eq. 5).
    Assumes fast dynamics are at quasi-stationary values; no error bound is provided.
  • ad hoc to paper The quadratic cost functional with lambda=60 captures the clinically appropriate trade-off between glucose reduction and exercise burden (Eq. 7).
    Lambda is calibrated by simulation; the cost is not derived from clinical utility data.
  • domain assumption The Topp model's hyperglycemic steady state at G=600 represents irreversible diabetes progression.
    Inherited from Ref. [18]; the control objective is to avoid this state and reach G=100.
invented entities (1)
  • Vl state variable (integral effect of IL-6 released during exercise)
    purpose: Mediates the long-term benefits of physical activity on insulin sensitivity (Eq. 1d) and beta-cell proliferation/apoptosis (Eqs. 2a-2d).
    No direct measurement of Vl is used; no falsifiable prediction outside the paper's own model is provided. The variable is inherited from Refs. [13,14].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Closed-loop robust control of long-term diabetes progression via physical activity management." pith.science (2026). https://pith.science/paper/V3O5FXCX

@misc{pith2026250112892,
  author       = {Pith},
  title        = {Pith review of: Closed-loop robust control of long-term diabetes progression via physical activity management},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V3O5FXCX}},
  note         = {Machine review of arXiv:2501.12892}
}
read the original abstract

Large clinical evidence acknowledges the crucial role played by physical activity in delaying the progression of type-2 diabetes. However, the literature lacks control approaches that leverage exercise for type-2 diabetes control and more in general lacks a quantitative assessment of medical guidelines on the recommended amount of physical activity to be performed, mainly due to the absence of mathematical models that suitably estimate its benefits on diabetes progression. In this work, in order to provide a control-theoretical formulation of the exercise, we design a feedback law in terms of recommended physical activity, following a model predictive control approach, based on a widespread compact diabetes progression model, suitably modified to properly account for the long-term effect of the exercise. Moreover we illustrate how the proposed approach proves to show reliable robustness properties with respect to initial conditions and parameter perturbations, which may be used to reflect inter-patient variability. Results are encouraging in view of the validation of the control law on comprehensive high-dimensional models of diabetes progression, with the aim of translating the prediction of the controller into reasonable recommendations and to quantitatively support medical decision-making.

Figures

Figures reproduced from arXiv: 2501.12892 by the authors.

Figure 1
Figure 1. Basal glucose concentration as a function of time in the op [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Equivalent control input ueq predicted by the MPC controller as a function of time, with the nominal parameters in [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Recommended duration of single exercise sessions as a fun [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Basal glucose concentration as a function of time in the co [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Beta-cell mass as a function of time in the controlled case c [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 29 canonical work pages

  1. [1]

    Closed-loop robust control of long-term diabetes progression v ia physi- cal activity management,

    P. De Paola, A. Borri, F. Dabbene, P. Palumbo, and A. Paglialonga, “Closed-loop robust control of long-term diabetes progression v ia physi- cal activity management,” in Submitted to Proceedings of the 11th IF AC Symposium on Robust Control Design (ROCOND’25) , 2025

    work page 2025

  2. [2]

    Global and societal implication s of the siabetes epidemic,

    P. Zimmet, G. Alberti, and J. Shaw, “Global and societal implication s of the siabetes epidemic,” Nature, vol. 414, no. 6865, p. 782, 2001

    work page 2001

  3. [3]

    Classification and diagnosis of dia betes mellitus and other categories of glucose intolerance,

    National Diabetes Data Group, “Classification and diagnosis of dia betes mellitus and other categories of glucose intolerance,” Diabetes, vol. 28, no. 12, pp. 1039–1057, 1979

    work page 1979

  4. [4]

    Applicability r esults of a nonlinear model-based robust blood glucose control algorithm,

    L. Kov´ acs, P. Szalay, Z. Alm´ assy, and L. Barkai, “Applicability r esults of a nonlinear model-based robust blood glucose control algorithm,” Journal of diabetes science and technology , vol. 7, no. 3, pp. 708–716, 2013

    work page 2013

  5. [5]

    Symbolic control design of an artificial pancreas for type-2 diabetes,

    A. Borri, G. Pola, P. Pepe, M. D. Di Benedetto, and P. Palumbo, “ Symbolic control design of an artificial pancreas for type-2 diabetes,” IEEE Transac- tions on Control Systems Technology , vol. 30, no. 5, pp. 2131–2146, 2021

    work page 2021

  6. [6]

    Model predictive control with integral action for artificial pancreas,

    G. P. Incremona, M. Messori, C. Toffanin, C. Cobelli, and L. Magni, “Model predictive control with integral action for artificial pancreas,” Control En- gineering Practice, vol. 77, pp. 86–94, 2018

    work page 2018

  7. [7]

    Individu alized model predictive control for the artificial pancreas: In silico evalu ation of closed-loop glucose control,

    M. Messori, G. Paolo Incremona, C. Cobelli, and L. Magni, “Individu alized model predictive control for the artificial pancreas: In silico evalu ation of closed-loop glucose control,” IEEE Control Systems Magazine , vol. 38, no. 1, pp. 86–104, 2018. 13

    work page 2018

  8. [8]

    Sampled-data observer-based glucose control for the artificia l pancreas,

    A. Borri, P. Palumbo, C. Manes, S. Panunzi, and A. De Gaetano, “Sampled-data observer-based glucose control for the artificia l pancreas,” Acta Polytech. Hungarica , vol. 14, no. 1, pp. 79–94, 2017

    work page 2017

Show all 29 references
  1. [9]

    Optimal periodic impulsive strategies in glycemic control,

    A. Borri, F. Cacace, A. De Gaetano, M. Pompa, and S. Panunzi, “ Optimal periodic impulsive strategies in glycemic control,” IEEE Transactions on Control Systems Technology, vol. 32, no. 6, pp. 2062–2074, 2024

    2024

  2. [10]

    Hemmingsen, G

    B. Hemmingsen, G. Gimenez-Perez, D. Mauricio, M. Roqu´ e i Figu ls, M. Metzendorf, and B. Richter, “Diet, physical activity or both fo r preven- tion or delay of type 2 diabetes mellitus and its associated complication s in people at increased risk of developing type 2 diabete...

    2017

  3. [11]

    Reduction in t he incidence of type 2 diabetes with lifestyle intervention or metformin,

    Diabetes Prevention Program Research Group, “Reduction in t he incidence of type 2 diabetes with lifestyle intervention or metformin,” New England journal of medicine , vol. 346, no. 6, pp. 393–403, 2002

    2002

  4. [12]

    The english national health service diabete s pre- vention programme (nhs dpp): A scoping review of existing evidence ,

    M. Whelan and L. Bell, “The english national health service diabete s pre- vention programme (nhs dpp): A scoping review of existing evidence ,” Diabetic Medicine, vol. 39, no. 7, p. e14855, 2022

    2022

  5. [13]

    The long-term effects of physical activity on blood glucos e reg- ulation: a model to unravel diabetes progression,

    P. F. De Paola, A. Paglialonga, P. Palumbo, K. Keshavjee, F. Dab bene, and A. Borri, “The long-term effects of physical activity on blood glucos e reg- ulation: a model to unravel diabetes progression,” IEEE Control Systems Letters, vol. 7, pp. 2916–2921, 2023

    2023

  6. [14]

    A Novel Mathematical Model for Predicting the Ben e- fits of Physical Activity on Type 2 Diabetes Progression,

    P. De Paola, A. Borri, F. Dabbene, K. Keshavjee, P. Palumbo, a nd A. Paglialonga, “A Novel Mathematical Model for Predicting the Ben e- fits of Physical Activity on Type 2 Diabetes Progression,” arXiv preprint , 2024

    2024

  7. [15]

    A model-based approach for glucose control via physical activity,

    P. F. De Paola, A. Borri, A. Paglialonga, P. Palumbo, and F. Dabbe ne, “A model-based approach for glucose control via physical activity,” in Studies in Health Technology and Informatics , 2025. In press

    2025

  8. [16]

    Allg¨ ower and A

    F. Allg¨ ower and A. Zheng, Nonlinear model predictive control , vol. 26. Birkh¨ auser, 2012

    2012

  9. [17]

    E. F. Camacho and C. B. Alba, Model predictive control. Springer science & business media, 2013

    2013

  10. [18]

    A model of β -cell mass, insulin, and glucose kinetics: pathways to diabetes,

    B. Topp, K. Promislow, G. Devries, R. M. Miura, and D. T Finegood , “A model of β -cell mass, insulin, and glucose kinetics: pathways to diabetes,” Journal of theoretical biology , vol. 206, no. 4, pp. 605–619, 2000

    2000

  11. [19]

    World health organization 2020 guidelines on physical activity and sedentary beh aviour,

    F. C. Bull, S. S. Al-Ansari, S. Biddle, K. Borodulin, M. P. Buman, G. Car- don, C. Carty, J.-P. Chaput, S. Chastin, R. Chou, et al. , “World health organization 2020 guidelines on physical activity and sedentary beh aviour,” British journal of sports medicine , vol. 54, no. 24...

    2020

  12. [20]

    Update on the effects of physical a ctivity on insulin sensitivity in humans,

    S. R. Bird and J. A. Hawley, “Update on the effects of physical a ctivity on insulin sensitivity in humans,” BMJ open sport & exercise medicine , vol. 2, no. 1, p. e000143, 2017

    2017

  13. [21]

    Theory and applications of hva c control systems–a review of model predictive control (mpc),

    A. Afram and F. Janabi-Sharifi, “Theory and applications of hva c control systems–a review of model predictive control (mpc),” Building and Envi- ronment, vol. 72, pp. 343–355, 2014

    2014

  14. [22]

    Optimal control, mpc a nd mpc- like algorithms for wave energy systems: An overview,

    N. Faedo, S. Olaya, and J. V. Ringwood, “Optimal control, mpc a nd mpc- like algorithms for wave energy systems: An overview,” IF AC Journal of Systems and Control , vol. 1, pp. 37–56, 2017

    2017

  15. [23]

    Mathematical modeling o f lung mechanics and pressure-controlled ventilation design for barotra uma mini- mization: A numerical simulation study,

    L. D’Orsi, A. Borri, and A. De Gaetano, “Mathematical modeling o f lung mechanics and pressure-controlled ventilation design for barotra uma mini- mization: A numerical simulation study,” International Journal of Robust and Nonlinear Control , 2024

    2024

  16. [24]

    Dose–response relationship between device-measured physical activity and incident type 2 diabetes: findings from the uk biobank prospective c ohort study,

    J. Boonpor, S. Parra-Soto, F. Petermann-Rocha, N. Lynsk ey, V. Cabanas- S´ anchez, N. Sattar, J. M. Gill, P. Welsh, J. P. Pell, S. R. Gray, et al. , “Dose–response relationship between device-measured physical activity and incident type 2 diabetes: findings from the uk biobank ...

    2023

  17. [25]

    The long-term effect of lifestyle interventions to prevent diabetes in the china da qing diabetes prevention study: a 2 0-year follow-up study,

    G. Li, P. Zhang, J. Wang, E. W. Gregg, W. Yang, Q. Gong, H. Li, H . Li, Y. Jiang, Y. An, et al. , “The long-term effect of lifestyle interventions to prevent diabetes in the china da qing diabetes prevention study: a 2 0-year follow-up study,” The Lancet, vol. 371, no. 9626, pp...

    2008

  18. [26]

    Long-term improvement in ins ulin sensitivity by changing lifestyles of people with impaired glucose tolera nce: 4-year results from the finnish diabetes prevention study,

    M. Uusitupa, V. Lindi, A. Louheranta, T. Salopuro, J. Lindstr¨ om, J. Tuomilehto, and F. D. P. S. Group, “Long-term improvement in ins ulin sensitivity by changing lifestyles of people with impaired glucose tolera nce: 4-year results from the finnish diabetes prevention study,”...

    2003

  19. [27]

    Preservation of β -cell function: the key to diabetes prevention,

    R. A. DeFronzo and M. A. Abdul-Ghani, “Preservation of β -cell function: the key to diabetes prevention,” The Journal of Clinical Endocrinology & Metabolism, vol. 96, no. 8, pp. 2354–2366, 2011

    2011

  20. [28]

    Mathematical models of diabetes prog ression,

    A. De Gaetano, T. Hardy, B. Beck, E. Abu-Raddad, P. Palumbo , J. Bue- Valleskey, and N. Pørksen, “Mathematical models of diabetes prog ression,” American Journal of Physiology-Endocrinology and Metabol ism, vol. 295, no. 6, pp. E1462–E1479, 2008

    2008

  21. [29]

    A mathematical model o f the pathogenesis, prevention, and reversal of type 2 diabetes,

    J. Ha, L. S. Satin, and A. S. Sherman, “A mathematical model o f the pathogenesis, prevention, and reversal of type 2 diabetes,” Endocrinology, vol. 157, no. 2, pp. 624–635, 2016. 15

    2016

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.