Identification and validation of an explainable prediction model of favorable outcome under integrative medicine treatment exposure in DKD adult patients: a retrospective cohort study - Takeaways - MDSpire

Identification and validation of an explainable prediction model of favorable outcome under integrative medicine treatment exposure in DKD adult patients: a retrospective cohort study

  • By

  • Li Jiang

  • Haojun Zhang

  • Yanmei Wang

  • Meihua Yan

  • Xiai Wu

  • July 16, 2026

Share

  • 1

    A predictive model for favorable outcomes in adult diabetic kidney disease (DKD) patients receiving integrative medicine treatment was developed and validated.

  • 2

    The study analyzed a cohort of 7,400 DKD patients, using a training set and a validation cohort to assess model performance.

  • 3

    XGBoost was identified as the optimal predictive model, achieving an AUC of 0.783 in training and 0.762 in the validation set.

  • 4

    Ten key variables were determined to be significant predictors of favorable outcomes, including creatinine, age, and uric acid levels.

  • 5

    An interactive web application was created to provide real-time predictions of treatment outcomes for DKD patients based on laboratory values.

Original Source(s)

Identification and validation of an explainable prediction model of favorable outcome under integrative medicine treatment exposure in DKD adult patients: a retrospective cohort study

  • by Li Jiang, Haojun Zhang, Yanmei Wang, Meihua Yan, Xiai Wu

  • July 16, 2026

Related Content

Prevalence of neurocognitive dysfunction and associated risk factors among follow-up diabetic type 2 patients in diabetes center of Al-Thawra hospital, Sana’a, Yemen

Ocular manifestations and immunological profiles of thyroid eye disease with lacrimal gland enlargement

Gender-Diverse Diagnostic Guidance Issued

ADLM releases new US recommendations to improve laboratory test interpretation and diagnostic decision-making for gender-diverse patients

  • July 15, 2026

  • 2 min