EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predicting 10-Year Diabetes Risk Through Physiological Acceleration: A Longitudinal Deep Learning Ensemble Approach

Sangsoo Kim iD, Seonghee Park iD, Jinmi Kim iD, Ha Jin Park iD, Soree Ryang iD, Myungsoo Im iD, Doohwa Kim iD, Kyeongjun Lee iD

Diagnostics · 25.03.2026 · 16 (7) · 992

DOI: 10.3390/diagnostics16070992 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 25.03.2026. Verfasst von Sangsoo Kim und 7 weiteren Autor:innen. Beteiligte Einrichtung: Division of Endocrinology and Metabolism, Department of Internal Medicine, Pusan National University Hospital, Busan 49241, Gyeongsangnam-do, Republi…. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Health and Welfare, Korea Health Industry Development Institute (KHIDI) https://ror.org/00fdzyk40, Ministry of Health and Welfare https://ror.org/00vxgjw72.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: Type 2 diabetes (T2D) develops gradually over many years through a prolonged preclinical phase, yet traditional static risk scores often fail to capture these dynamic metabolic trajectories. We propose a longitudinal deep learning framework to predict the 10-year risk of Type 2 diabetes onset defined by comprehensive ADA criteria by modeling the physiological acceleration of routine clinical biomarkers.

Methods

Utilizing an 18-year longitudinal dataset from the community-based Korean Genome and Epidemiology Study (KoGES) cohort, we selected N=4354 participants with complete follow-up records, ensuring high data integrity without requiring synthetic data augmentation. We constructed a 3-dimensional tensor of 21 non-invasive clinical variables spanning a 6-year observation window. To resolve the inherent precision-recall trade-offs of individual models, we developed a stacking ensemble that integrates Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures via a logistic regression meta-learner. To evaluate the added value of longitudinal modeling, we compared this dynamic framework against a static XGBoost baseline that only saw the most recent data.

Results

Evaluated on an independent test set (n=874), the ensemble significantly outperformed baseline models, achieving an overall accuracy of 0.90 (95% CI: 0.88–0.92) and an AUROC of 0.94 (95% CI: 0.93–0.95). By harmonizing LSTM’s sensitivity and GRU’s precision, the model yielded an exceptional Positive Predictive Value (PPV) of 0.97, a sensitivity of 0.80, and a specificity of 0.98.

Conclusions

This framework provides a highly accurate, resource-efficient triage instrument for T2D screening, thereby reducing unnecessary clinical alerts and improving screening efficiency.

Schlagwörter

Stichwörter der Autor:innen

Type 2 Diabetes Mellitusdeep learningensemble learninglongitudinal trajectoryprecision medicine

Autor:innen und Institutionen

  • Division of Endocrinology and Metabolism, Department of Internal Medicine, Pusan National University Hospital, Busan 49241, Gyeongsangnam-do, Republic of Korea.
  • Biomedical Research Institute, Pusan National University Hospital, Busan 49241, Gyeongsangnam-do, Republic of Korea.
  • Department of Internal Medicine, Pusan National University School of Medicine, Busan 49241, Gyeongsangnam-do, Republic of Korea.
  • Department of Mathematics and Big Data Science, Kumoh National Institute of Technology, Gumi 39177, Gyeongsangbuk-do, Republic of Korea.
  • Department of Biostatistics, Clinical Trial Center, Biomedical Research Institute, Pusan National University Hospital, Busan 49241, Gyeongsangnam-do, Republic of Korea.

ORCID-Profile: Sangsoo Kim, Seonghee Park, Jinmi Kim, Ha Jin Park, Soree Ryang, Myungsoo Im, Doohwa Kim, Kyeongjun Lee

Förderung

  • Ministry of Health and Welfare
  • Korea Health Industry Development Institute (KHIDI) https://ror.org/00fdzyk40
  • Ministry of Health and Welfare https://ror.org/00vxgjw72

Zitieren

APA

Kim, S., Park, S., Kim, J., Park, H. J., Ryang, S., Im, M., Kim, D., & Lee, K. (2026). Predicting 10-Year Diabetes Risk Through Physiological Acceleration: A Longitudinal Deep Learning Ensemble Approach. Diagnostics, 16(7), 992. https://doi.org/10.3390/diagnostics16070992

Vancouver

Kim S, Park S, Kim J, Park HJ, Ryang S, Im M, et al. Predicting 10-Year Diabetes Risk Through Physiological Acceleration: A Longitudinal Deep Learning Ensemble Approach. Diagnostics. 2026;16(7):992. doi:10.3390/diagnostics16070992

BibTeX
@article{Kim2026,
  author = {Kim, Sangsoo and Park, Seonghee and Kim, Jinmi and Park, Ha Jin and Ryang, Soree and Im, Myungsoo and Kim, Doohwa and Lee, Kyeongjun},
  title = {Predicting 10-Year Diabetes Risk Through Physiological Acceleration: A Longitudinal Deep Learning Ensemble Approach},
  journal = {Diagnostics},
  year = {2026},
  volume = {16},
  number = {7},
  pages = {992},
  doi = {10.3390/diagnostics16070992}
}
Kostenlos starten →