EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Diagnostic model based on machine learning with multi‐feature analysis for diabetic peripheral neuropathy in type 2 diabetes: A single‐center retrospective study

Diagnosemodell auf Basis von Machine Learning mit Multi-Feature-Analyse für diabetische periphere Neuropathie bei Typ-2-Diabetes: singlezentrische retrospektive Studie

Xi Zheng iD, Keke Yang, Ruyi Qu, Xiaomin Peng, Ling Yang

Journal of Diabetes Investigation · 30.10.2026

DOI: 10.1111/jdi.70473 ↗

Auf einen Blick

Bei 770 Patient:innen mit Typ-2-Diabetes wurden ML-Modelle zur Vorhersage einer diabetischen peripheren Neuropathie entwickelt. Am besten schnitten logistische Regression, XGBoost und LGBM ab.

Fragestellung

Lassen sich ML-Modelle mit mehreren Merkmalen zur Vorhersage des DPN-Risikos bei Patient:innen mit Typ-2-Diabetes entwickeln?

Methode

Retrospektive Single-Center-Studie mit 770 Patient:innen; 91 Variablen; Training/Test 7:3; Merkmalsauswahl mit Boruta; 11 ML-Algorithmen; Bewertung mit AUC, Kalibrierungs- und Entscheidungsanalyse; SHAP.

Ergebnisse

26,62 % (205/770) hatten DPN. Die Modelle mit neurologischen Untersuchungen erzielten die beste Leistung. Als wichtigste Prädiktoren wurden u. a. Stimmgabeltest, Achillessehnenreflex und Temperaturempfinden identifiziert.

Schlussfolgerung der Autor:innen

Der ML-Ansatz ermöglichte diagnostische Modelle mit stabiler Leistung. SHAP-Methoden können die klinische Anwendbarkeit der Modelle verbessern.

Kernaussagen

  • 26,62 % (205/770) hatten DPN
  • Beste Modelle: LR, XGBoost und LGBM
  • Wichtige Prädiktoren: Stimmgabeltest, Reflexe, Temperaturempfinden

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Diabetes Investigation, online erschienen am 30.10.2026. Verfasst von Xi Zheng und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of General Practice, Jiangwan Town Subdistrict Community Health Service Center, Hongkou District, Shanghai, China. Die Arbeit nennt 65 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Diabetic peripheral neuropathy (DPN) is a common and serious complication of diabetes, which imposes a significant economic burden on both society and individuals. Therefore, early identification and diagnosis are crucial for the prevention and treatment of DPN. This study aims to develop machine learning (ML) models based on a variety of different characteristics to predict the risk of DPN in patients with type 2 diabetes mellitus.

Methods

We retrospectively reviewed the data of 770 patients screened for DPN with type 2 diabetes mellitus in the Department of General Practice, Jiangwan Town Subdistrict Community Health Service Center between May 2024 and July 2024. The data included 91 variables for baseline description analysis. The dataset was divided into a training and test set for the model evaluation in a 7:3 ratio, and feature selection was performed with the Boruta algorithm. Eleven ML algorithms were used to develop diagnostic models for DPN. The area under the receiver operating characteristic curve, calibration curves, and decision curves analysis were applied to evaluate the performance of the models, and the Shapley Additive Explanations (SHAP) method was adopted to interpret the performance of the models and the importance of features.

Results

The study included 770 type 2 diabetes mellitus inpatients, with a mean age of 70.88 ± 5.88 years, and 26.62% (205/770) had DPN. Integrated with the results of evaluation, we selected logistic regression (LR), extreme gradient boosting (XGBoost), and light gradient boosting machine (LGBM) models including neurological examinations as the top‐performing algorithms. Based on variable importance ranking from the three models with overlap analysis, we identified 12 key predictors. Among the three models, the tuning fork test, ankle reflex, and temperature sensation test were all identified as the top three important predictors. Meanwhile, the low‐density lipoprotein cholesterol (LDL‐C) in the XGBoost model and the total cholesterol (Tch) in the LGBM were the important features for predicting DPN. In addition, excluding neurological examination findings, both LDL‐C and Tch were important indicators with high contribution degree.

Conclusions

The ML approach helped establish diagnostic models with stable performance. Meanwhile, the use of interpretable SHAP methods can enhance the clinical applicability of the models.

Schlagwörter

Stichwörter der Autor:innen

Diabetic peripheral neuropathyRisk factorsmachine learning

Autor:innen und Institutionen

  • Department of General Practice, Jiangwan Town Subdistrict Community Health Service Center, Hongkou District, Shanghai, China.
  • Department of Geriatrics, Shanghai Fourth People's Hospital Affiliated to Tongji University, Shanghai, China.

ORCID-Profile: Xi Zheng

Zitieren

APA

Zheng, X., Yang, K., Qu, R., Peng, X., & Yang, L. (2026). Diagnostic model based on machine learning with multi‐feature analysis for diabetic peripheral neuropathy in type 2 diabetes: A single‐center retrospective study. Journal of Diabetes Investigation. https://doi.org/10.1111/jdi.70473

Vancouver

Zheng X, Yang K, Qu R, Peng X, Yang L. Diagnostic model based on machine learning with multi‐feature analysis for diabetic peripheral neuropathy in type 2 diabetes: A single‐center retrospective study. Journal of Diabetes Investigation. 2026. doi:10.1111/jdi.70473

BibTeX
@article{Zheng2026,
  author = {Zheng, Xi and Yang, Keke and Qu, Ruyi and Peng, Xiaomin and Yang, Ling},
  title = {Diagnostic model based on machine learning with multi‐feature analysis for diabetic peripheral neuropathy in type 2 diabetes: A single‐center retrospective study},
  journal = {Journal of Diabetes Investigation},
  year = {2026},
  doi = {10.1111/jdi.70473}
}
Kostenlos starten →