A Dynamic Bayesian Network model for long-term simulation of clinical complications in type 1 diabetes
Open accessAbstract
The increasing prevalence of diabetes and its related complications is raising the need for effective methods to predict patient evolution and for stratifying cohorts in terms of risk of developing diabetes-related complications. In this paper, we present a novel approach to the simulation of a type 1 diabetes population, based on Dynamic Bayesian Networks, which combines literature knowledge with data mining of a rich longitudinal cohort of type 1 diabetes patients, the DCCT/EDIC study. In particular, in our approach we simulate the patient health state and complications through discretized variables. Two types of models are presented, one entirely learned from the data and the other partially driven by literature derived knowledge. The whole cohort is simulated for fifteen years, and the simulation error (i.e. for each variable, the percentage of patients predicted in the wrong state) is calculated every year on independent test data. For each variable, the population predicted in the wrong state is below 10% on both models over time. Furthermore, the distributions of real vs. simulated patients greatly overlap. Thus, the proposed models are viable tools to support decision making in type 1 diabetes.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Several studies utilized risk factors (mainly from EHRs) to develop AI algorithms to predict various types of diabetes-related complications with favorable predictive performance.
“In the prediction of diabetic complications, several studies tried to utilize risk factors (mainly from EHRs) to develop AI algorithms to predict various types of diabetes-related complications with favorable predictive performance.94,95,96,97,98,99,100”
From Artificial intelligence in diabetes management: Advancements, opportunities, and challenges (2023) · cites this paper as [95]
The Discussion reports that dynamic Bayesian networks modeled two type 1 diabetes complications, nephropathy and cardiovascular disease, with less than 10% error in population-state trajectories on independent test data. However, §2.1 identifies the data as coming from the DCCT and EDIC longitudinal studies, not EHRs, and the reported performance concerns cohort simulation rather than demonstrated individual-level complication prediction. Page numbers are unavailable in our copy. (LLM, conf=0.85)
Experimental — accuracy not yet measuredMarini et al. reported that a BN using risk factors to estimate long-term development and progression of complications in T1D predicted the wrong state in below 10% of the population on both DDO-DBN and EI-DBN.
“the population predicted in the wrong state was below 10% on both DDO-DBN and EI-DBN”
From Artificial intelligence in diabetes management: Advancements, opportunities, and challenges (2023) · cites this paper as Marini et al.95
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Fox, C. S., Pencina, M. J., Wilson, P. W. F., Paynter, N. P., Vasan, R. S., & D’Agostino, R. B. (2008) · Diabetes Care · doi:10.2337/dc08-0025 · Cited by 247 · see citing works
Flagged first, then by citations
Research works printed but not linked: 8
- J.E. Tarride, R. Hopkins, G. Blackhouse, J.M. Bowen, M. Bischof, C. Von Keyserlingk, D. O'Reilly, F. Xie, R. Goeree, A review of methods used in long- term cost-effectiveness models of diabetes mellitus treatment, Pharmacoeconomics 28 (4) (2010) 255-277.
- F.V. Jensen, T.D. Nielsen, Bayesian Networks and Decision Graphs, second ed., Springer, 2007.
- U. Kjrulff, A computational scheme for reasoning in dynamic probabilistic networks, in: Proceedings of the Eighth Conference on Uncertainty in Artificial Intelligence, Morgan Kaufman, 1992, pp. 121-129.
- K.P. Murphy, Dynamic Bayesian Networks: Representation, Inference and Learning, PhD thesis, 2002.
- K. Exarchos, C. Carpeggiani, G. Rigas, T. Exarchos, F. Vozzi, A. Sakellarios, et al., A multiscale approach for modeling atherosclerosis progression, IEEE J. Biomed. Health Inform. 19 (2) (2015) 709-719.
- F. Glover, M. Laguna, Tabu Search, vol. 22, Kluwer academic publishers, Boston, 1997.
- R. Schafer, W. Thomas, Assessing temporally variable user properties with dynamic Bayesian networks, in: User Modeling, Springer, Vienna, 1997, pp. 377-388.
- W.J. Conover, Practical Nonparametric Statistics, John Wiley & Sons, New York, 1971. pp. 295-301 (one-sample Kolmogorov test), and 309-314 (two-sample Smirnov test).
Cited by
Sorted by citations
Artificial intelligence in diabetes management: Advancements, opportunities, and challenges
Guan, Z., Li, H., Liu, R., Cai, C., Liu, Y., Li, J., Wang, X., Huang, S., Wu, L., Liu, D., et al. (2023) · Cell Reports Medicine · doi:10.1016/j.xcrm.2023.101213 · Cited by 287 · see citing works
- Authors
- Marini, S., Trifoglio, E., Barbarini, N., Sambo, F., Di Camillo, B., Malovini, A., Manfrini, M., Cobelli, C., & Bellazzi, R.
- Affiliations
- University of Pavia (IT)
- University of Padua (IT)
- Istituti Clinici Scientifici Maugeri (IT)
- Corresponding author
- Marini, Simone
- ORCID
- Venue
- Journal of Biomedical Informatics
- Publisher
- Elsevier BV
- Published
- 2015-10
- Volume
- 57
- Pages
- 369-376
- ISSN
- 1532-0464 (ISSN-L)
- OpenAlex
- W1257817829
- Cited by
- 61
- References
- 1
- Citations by year
- 2026: 1 · 2025: 5 · 2024: 4 · 2023: 6 · 2022: 9 · 2021: 10 · 2020: 7 · 2019: 6 · 2018: 9 · 2016: 3 · 2015: 1
- FWCI
- 8.35
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- Locations
- Journal of Biomedical Informatics (publishedVersion)
- PubMed (publishedVersion)
- Padua Research Archive (University of Padova) (submittedVersion)
- Topics
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Keywords
- type 1 diabetes, diabetes complications, dynamic Bayesian networks, population simulation, longitudinal cohort, risk stratification, clinical decision support, simulation error, knowledge-based modeling
- MeSH terms
- Bayes Theorem, Computer Simulation, Diabetes Mellitus, Type 1, Humans, Diabetes Complications, Data Mining
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.
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
