CoChat

A comparison among three maximal mathematical models of the glucose-insulin system

Open access
Pompa, Marcello, Panunzi, Simona, Borri, Alessandro, De Gaetano, Andrea · 2021 · PLOS ONE

The DOI matches its Crossref record; 28 of the 31 references it cites were found in a registry or an index; 17 of the 68 sources its claims cite were checked for support.

Abstract

The most well-known and widely used mathematical representations of the physiology of a diabetic individual are the Sorensen and Hovorka models as well as the UVAPadova Simulator. While the Hovorka model and the UVAPadova Simulator only describe the glucose metabolism of a subject with type 1 diabetes, the Sorensen model was formulated to simulate the behaviour of both normal and diabetic individuals. The UVAPadova model is the most known model, accepted by the FDA, with a high level of complexity. The Hovorka model is the simplest of the three models, well documented and used primarily for the development of control algorithms. The Sorensen model is the most complete, even though some modifications were required both to the model equations (adding useful compartments for modelling subcutaneous insulin delivery) and to the parameter values. In the present work several simulated experiments, such as IVGTTs and OGTTs, were used as tools to compare the three formulations in order to establish to what extent increasing complexity translates into richer and more correct physiological behaviour. All the equations and parameters used for carrying out the simulations are provided.

Overview

CoChat Summary

TL;DR

This paper compares three maximal glucose-insulin models—Sorensen, Hovorka, and UVAPadova S2017—across simulated IVGTT and OGTT protocols and an OGTT fit to normal-subject data. It finds that the original Sorensen model underestimates diabetic hyperglycemia unless insulin-sensitivity parameters are modified, after which all three models behave similarly, though Sorensen uniquely captures the post-hypoglycemic rebound.

Key findings

  • In an in-silico OGTT plus insulin bolus, the original Sorensen model reached a maximum glycemia of 10 mM versus 18 mM for Hovorka and UVAPadova, and returned to baseline in about 300 min, roughly half the time of the other two models.
  • After modifying Sorensen parameters—e.g., FPIC 0.150→0.240, FLIC 0.400→0.459, peripheral venous insulin volume 0.740→2.442, and insulin-dependent glucose utilization parameters decreased about tenfold—its predictions became comparable to Hovorka and UVAPadova.
  • Fitting to normal-subject OGTT data, Sorensen gave the best fit and was the only model predicting the rebound around minute 200; estimated insulin boluses were 56387.7 pmol (~8 IU) for Sorensen, 2361 mU (~2.3 IU) for Hovorka, and 849 pmol/kg (~8 IU) for UVAPadova. UVAPadova required 10 free parameters versus 4 for Sorensen and 3 for Hovorka, and all UVAPadova free-parameter CVs were >100%.
  • In three successive OGTTs in one day, Sorensen reproduced the same glycemic pattern each time; UVAPadova drove glucose progressively lower (below 2 mM), while Hovorka predicted lower glucose in the second and third OGTTs but never below 3 mM.

Methods

  • Four in-silico experiments were simulated: IVGTT (0.5 g/kg glucose over 3 min plus basal insulin 6.67 mU/min), IVGTT plus 1000 mU insulin bolus, OGTT (100 g glucose over 1 min) plus basal insulin, and OGTT plus basal insulin plus 1000 mU bolus; initial steady state was glucose 5 mmol/L with basal insulin 6.67 mU/min.
  • Models compared were the Sorensen model (revised with gastrointestinal tract and subcutaneous insulin compartments, pancreatic insulin secretion removed for T1DM), the Hovorka model, and the UVAPadova S2017 model; all equations and parameters used are provided.
  • For OGTT fitting, weighted least squares was applied to observed glucose concentrations from a normal individual (Sorensen’s PhD thesis) with 100 g glucose, estimating bolus insulin and selected parameters; these estimates were then used to simulate three OGTTs in one day.

Limitations

  • Stated: the Sorensen T1DM adaptation assumes diabetic physiology is otherwise the same as normal apart from absent insulin production, so it omits diabetic abnormalities; no actual diabetic-subject perturbation data were available, so it cannot be determined which model is physiologically more credible; OGTT fitting used normal-subject data because OGTT is not standard in diabetic patients.
  • Stated: UVAPadova has missing parameter values and some equations described only qualitatively (kp1 and kir were kept constant); from purely in-silico experiments, definitive conclusions cannot be drawn.
  • Evident: Sorensen parameter changes were trial-and-error adjustments to match the other models rather than estimates from diabetic data; UVAPadova free parameters all had CV >100%, indicating poor identifiability; the paper also notes that Hovorka and UVAPadova EGP formulations can predict negative EGP at high insulin.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0257789

Cites

28 of 28 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig 1 Blood glucose concentrations from Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models with the original Sorensen model, during an in-silico OGTT experiment in conjunction with an insulin bolus.
Fig 1 Blood glucose concentrations from Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models with the original Sorensen model, during an in-silico OGTT experiment in conjunction with an insulin bolus.CC BY · PubMed Central
Fig 2 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models with the modified Sorensen model, during an in-silico OGTT experiment in conjunction with an insulin bolus.
Fig 2 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models with the modified Sorensen model, during an in-silico OGTT experiment in conjunction with an insulin bolus.CC BY · PubMed Central
Fig 3 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models, during the in-silico OGTT experiment.
Fig 3 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models, during the in-silico OGTT experiment.CC BY · PubMed Central
Fig 4 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models, during the in-silico IVGTT experiment in conjunction with an insulin bolus.
Fig 4 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models, during the in-silico IVGTT experiment in conjunction with an insulin bolus.CC BY · PubMed Central
Fig 5 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models, during the in-silico IVGTT experiment.
Fig 5 Blood glucose concentrations from the Hovorka (solid line), Sorensen (dashed line) and UVAPadova (dotted line) models, during the in-silico IVGTT experiment.CC BY · PubMed Central
Fig 6 A)Blood glucose concentrations of Sorensen (dashed line), Hovorka (solid line), UVAPadova (dashed dotted line) models with data points from Sorensen PhD thesis (asterisks); B) Hovorka (solid line) vs data points (asterisks); C) Sorensen (dashed line) vs data points (asterisks); D) UVAPadova (dashed dotted line) vs data points (asterisks).
Fig 6 A)Blood glucose concentrations of Sorensen (dashed line), Hovorka (solid line), UVAPadova (dashed dotted line) models with data points from Sorensen PhD thesis (asterisks); B) Hovorka (solid line) vs data points (asterisks); C) Sorensen (dashed line) vs data points (asterisks); D) UVAPadova (dashed dotted line) vs data points (asterisks).CC BY · PubMed Central
Fig 7 A)Blood glucose concentrations of Sorensen (dashed line), Hovorka (solid line), UVAPadova (dashed dotted line) models after three OGTT in conjunction with three insulin bolus; B) Hovorka (solid line); C) Sorensen (dashed line); UVAPadova (dashed dotted line).
Fig 7 A)Blood glucose concentrations of Sorensen (dashed line), Hovorka (solid line), UVAPadova (dashed dotted line) models after three OGTT in conjunction with three insulin bolus; B) Hovorka (solid line); C) Sorensen (dashed line); UVAPadova (dashed dotted line).CC BY · PubMed Central

Verification

Claims the paper makes

9 Valid · 1 Not valid · 18 Not checked · 67 Own results and interpretations

44 more claims

Create a free account to see all 95 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 31
  • Probably not a substantial claim. 21

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

3 of 3

Citations

Cites

Research works printed but not linked: 5

  1. C Guyton, 2006, Textbook of Medical Physiology Eleventh Edition
  2. WHO, 2016, Gobal report on diabetes
  3. HHS, 2020, National Diabetes Statistics Report 2020 Estimates of Diabetes and Its Burden in the United States
  4. JT Sorensen, 1985, A Physiologic Model of Glucose Metabolism in Man and Its Use to Design and Improved Insulin Therapies for Diabetes, Massachussets Institute of Technology
  5. L Hinshaw, 2013, Pathogenesis of NIDDM. A balanced overview, Diabetes, 62, 2223

Cited by

See all 24

Metadata

Authors
Pompa, M., Panunzi, S., Borri, A., & De Gaetano, A.
Affiliations
  • Università Cattolica del Sacro Cuore (IT)
  • National Research Council (IT)
Corresponding author
  • Pompa, Marcello
  • Panunzi, Simona
  • Borri, Alessandro
Venue
PLOS ONE
Published
2021-09-27
Volume
16
Issue
9
Pages
e0257789
ISSN
1932-6203 (ISSN-L)
OpenAlex
W3202301434
Cited by
24
References
31
Citations by year
2026: 3 · 2025: 6 · 2024: 7 · 2023: 5 · 2022: 3
FWCI
2.79
Type
journal article
Language
en
Open access
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Management and Research
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Pancreatic function and diabetes
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Treatment and Management
Keywords
glucose-insulin system, mathematical model, Hovorka model, type 1 diabetes, model comparison, model complexity, subcutaneous insulin administration, intravenous glucose tolerance test, oral glucose tolerance test
MeSH terms
Computer Simulation, Diabetes Mellitus, Glucose, Humans, Insulin, Models, Biological

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.