CoChat

Machine Learning Models for Blood Glucose Level Prediction in Patients With Diabetes Mellitus: Systematic Review and Network Meta-Analysis

Liu, Kui, Li, Linyi, Ma, Yifei, Jiang, Jun, Liu, Zhenhua, Ye, Zichen, Liu, Shuang, Pu, Chen, Chen, Changsheng, Wan, Yi · 2023 · JMIR Medical Informatics

Published version of a preprint

Cited by 67

The DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.

Abstract

Background: Machine learning (ML) models provide more choices to patients with diabetes mellitus (DM) to more properly manage blood glucose (BG) levels. However, because of numerous types of ML algorithms, choosing an appropriate model is vitally important. Objective: In a systematic review and network meta-analysis, this study aimed to comprehensively assess the performance of ML models in predicting BG levels. In addition, we assessed ML models used to detect and predict adverse BG (hypoglycemia) events by calculating pooled estimates of sensitivity and specificity. Methods: PubMed, Embase, Web of Science, and Institute of Electrical and Electronics Engineers Explore databases were systematically searched for studies on predicting BG levels and predicting or detecting adverse BG events using ML models, from inception to November 2022. Studies that assessed the performance of different ML models in predicting or detecting BG levels or adverse BG events of patients with DM were included. Studies with no derivation or performance metrics of ML models were excluded. The Quality Assessment of Diagnostic Accuracy Studies tool was applied to assess the quality of included studies. Primary outcomes were the relative ranking of ML models for predicting BG levels in different prediction horizons (PHs) and pooled estimates of the sensitivity and specificity of ML models in detecting or predicting adverse BG events. Results: In total, 46 eligible studies were included for meta-analysis. Regarding ML models for predicting BG levels, the means of the absolute root mean square error (RMSE) in a PH of 15, 30, 45, and 60 minutes were 18.88 (SD 19.71), 21.40 (SD 12.56), 21.27 (SD 5.17), and 30.01 (SD 7.23) mg/dL, respectively. The neural network model (NNM) showed the highest relative performance in different PHs. Furthermore, the pooled estimates of the positive likelihood ratio and the negative likelihood ratio of ML models were 8.3 (95% CI 5.7-12.0) and 0.31 (95% CI 0.22-0.44), respectively, for predicting hypoglycemia and 2.4 (95% CI 1.6-3.7) and 0.37 (95% CI 0.29-0.46), respectively, for detecting hypoglycemia. Conclusions: Statistically significant high heterogeneity was detected in all subgroups, with different sources of heterogeneity. For predicting precise BG levels, the RMSE increases with a rise in the PH, and the NNM shows the highest relative performance among all the ML models. Meanwhile, current ML models have sufficient ability to predict adverse BG events, while their ability to detect adverse BG events needs to be enhanced. Trial Registration: PROSPERO CRD42022375250; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=375250

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.2196/47833

Cites

10 of 10 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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?

Self-check

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

The self-check needs the paper’s full text.

Citations

Cites

Show all 10 references

Research works printed but not linked: 62

  1. 10.1186/s12916-022-02709-2 · doi:10.1186/s12916-022-02709-2
  2. 10.1016/j.diabres.2021.109040 · DOI
  3. 10.1080/0886022X.2020.1866010 · DOI
  4. 10.4093/dmj.2020.0100 · DOI
  5. 10.1007/s11892-019-1189-3 · DOI · Paper page
  6. Jain, AMC, 2012, J Diab Foot Comp, 4, 1
  7. 10.1111/j.1728-4465.1999.00067.x · DOI
  8. 10.2196/14340 · DOI
  9. 10.1001/jamainternmed.2014.1824 · DOI · Paper page
  10. 10.1080/0886022X.2022.2056053 · DOI · Paper page
  11. 10.1016/j.artmed.2021.102120 · DOI
  12. 10.1186/s12911-019-0943-4 · DOI
  13. 10.1109/EMBC.2013.6610766 · DOI
  14. 10.3390/app11041742 · DOI
  15. 10.1089/dia.2013.0104 · DOI · Paper page
  16. 10.1109/EMBC.2014.6943589 · DOI
  17. 10.1089/dia.2009.0076 · DOI · Paper page
  18. 10.3390/s21051647 · DOI · Paper page
  19. 10.1007/s41666-020-00068-2 · DOI
  20. 10.1371/journal.pmed.1000097 · DOI · Paper page
  21. 10.1136/bmj.b2700 · DOI · Paper page
  22. 10.1136/bmj.327.7414.557 · DOI · Paper page
  23. 10.3390/s22041665 · DOI
  24. 10.1111/dme.13409 · DOI
  25. 10.3390/s20061705 · DOI
  26. 10.1016/j.diabres.2020.108388 · DOI
  27. 10.1089/dia.2019.0458 · DOI
  28. 10.2337/dc19-1743 · DOI · Paper page
  29. 10.1109/JBHI.2019.2938305 · DOI
  30. 10.1177/1932296819868727 · DOI · Paper page
  31. 10.1016/j.ijmedinf.2019.03.008 · DOI
  32. 10.1177/1932296819880864 · DOI
  33. 10.1016/j.eswa.2011.02.020 · DOI
  34. 10.1109/IEMBS.2010.5627233 · DOI
  35. 10.3390/e22010081 · DOI
  36. 10.2196/11990 · DOI
  37. 10.1089/dia.2013.0069 · DOI · Paper page
  38. 10.1177/193229681000400109 · DOI
  39. 10.1007/BF02344732 · DOI
  40. BertachiABiagiLContrerasILuoNVehíJPrediction of blood glucose levels and nocturnal hypoglycemia using physiological models and artificial neural networks20133rd International Workshop on Knowledge Discovery in Healthcare DataJuly 13, 2018Stockholm, Sweden
  41. Eljil, KAAS, 2014, Predicting Hypoglycemia in Diabetic Patients Using Machine Learning Techniques
  42. 10.1016/j.knosys.2020.106134 · DOI
  43. 10.1089/dia.2019.0435 · DOI
  44. 10.1109/jbhi.2019.2931842 · DOI · Paper page
  45. 10.1109/TBME.2012.2188893 · DOI · Paper page
  46. 10.1109/EMBC44109.2020.9176695 · DOI
  47. 10.1109/JBHI.2021.3100558 · DOI
  48. 10.1016/j.bbe.2020.10.004 · DOI
  49. 10.1177/1932296820922622 · DOI · Paper page
  50. 10.1002/dmrr.3348 · DOI
  51. 10.1177/1932296816670400 · DOI
  52. 10.1177/1932296814554260 · DOI · Paper page
  53. 10.1007/s10439-011-0446-7 · DOI
  54. 10.1109/TCYB.2013.2283296 · DOI · Paper page
  55. 10.1016/j.artmed.2012.04.003 · DOI
  56. 10.1109/EMBC.2012.6346724 · DOI
  57. 10.1109/JBHI.2019.2931782 · DOI
  58. 10.1109/EMBC.2018.8513069 · DOI
  59. 10.1109/ijcnn.2010.5596916 · DOI
  60. 10.1001/jama.271.9.703 · doi:10.1001/jama.271.9.703 · Paper page
  61. 10.2196/22458 · DOI · Paper page
  62. McShinskyRMarshallBComparison of forecasting algorithms for type 1 diabetic glucose prediction on 30 and 60-minute prediction horizons2020KDH@ECAI 2020: 5th International Workshop on Knowledge Discovery in Healthcare Data co-located with 24th European Conference on Artificial IntelligenceAugust 29-30, 2020Santiago de Compostela, Spain, and virtually

Cited by

Metadata

Authors
Liu, K., Li, L., Ma, Y., Jiang, J., Liu, Z., Ye, Z., Liu, S., Pu, C., Chen, C., & Wan, Y.
Venue
JMIR Medical Informatics
Published
2023-11-20
Volume
11
Pages
e47833
ISSN
2291-9694
Cited by
67
References
10
Language
en

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

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.