Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Predicting and explaining poor prognosis in diabetic kidney disease using SHAP-based interpretable machine learning.

Man Qian1, Lin Li1, Yanli Cheng1

  • 1Department of Nephrology, The First Hospital of Jilin University, Changchun, China.

Iscience
|June 15, 2026
PubMed
Summary

Related Concept Videos

Diabetic Nephropathy01:28

Diabetic Nephropathy

Definition Diabetic nephropathy is a chronic kidney complication that results from prolonged hyperglycemia.Prevalence It is the most common cause of chronic kidney disease (CKD) and end-stage renal disease (ESRD) worldwide, affecting up to half of individuals with diabetes.Pathophysiology • Sustained hyperglycemia triggers multiple hemodynamic and metabolic changes in the kidney. • Early in the disease, increased renal blood flow and glomerular hyperfiltration occur due to afferent arteriolar...
Diabetic Retinopathy01:27

Diabetic Retinopathy

DefinitionDiabetic retinopathy is a microvascular complication of diabetes affecting the retinal blood vessels.Risk FactorsDiabetic retinopathy is present in almost all individuals with type 1 diabetes and more than 60% of those with type 2 diabetes after two decades of disease.The risk increases with poor glycemic control, hypertension, dyslipidemia, smoking, pregnancy, and puberty.Although cataracts and glaucoma are also more frequent in people with diabetes, retinopathy remains the leading...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Association of homocysteine, asymmetric dimethylarginine, and nitric oxide with preeclampsia.

Archives of gynecology and obstetrics·2009
Same author

[Measurement of fine-structure branching ratios for Rb-He optical collisions].

Guang pu xue yu guang pu fen xi = Guang pu·2009
Same author

[The prognostic value of a modified WPSS based on routine laboratory parameters in patients with myelodysplastic syndromes: a preliminary result].

Zhonghua xue ye xue za zhi = Zhonghua xueyexue zazhi·2009
Same author

Lessons learned from the application of a Vietnamese surname list for survey research.

Journal of immigrant and minority health·2009
Same author

Variations of protein levels in human amniotic fluid stem cells CD117/2 over passages 5-25.

Journal of proteome research·2009
Same author

[Effects of Shenwu capsule and its component tetrahydroxystilbene glucoside on expression of neurotrophic factors in lumbar spinal cord of aged rats].

Zhongguo Zhong yao za zhi = Zhongguo zhongyao zazhi = China journal of Chinese materia medica·2009

Machine learning models accurately predict outcomes for diabetic kidney disease (DKD) patients. Key factors like eGFR and C3 levels offer personalized prognostic insights for better DKD management.

Area of Science:

  • Nephrology
  • Medical Informatics
  • Biostatistics

Background:

  • Prognostic assessment is crucial for personalized management of diabetic kidney disease (DKD).
  • Developing accurate predictive models can improve patient outcomes and healthcare resource allocation.
  • Existing methods may not fully capture the complex interplay of factors influencing DKD progression.

Purpose of the Study:

  • To develop and validate machine learning (ML) models for predicting a composite endpoint in biopsy-proven diabetic kidney disease (DKD) patients.
  • To identify key clinical and biochemical features that are most predictive of DKD progression.
  • To assess the generalizability and robustness of the developed ML models.

Main Methods:

  • Eight machine learning models were developed using data from 180 biopsy-proven DKD patients.
Keywords:
Artificial intelligenceHealth sciencesMachine learningNephrology

Related Experiment Videos

  • Models predicted a composite endpoint including all-cause mortality, dialysis initiation, or renal transplantation.
  • External validation and SHAP (SHapley Additive exPlanations) analysis were employed to assess generalizability and feature importance.

    • Main Results:

    • The Naive Bayes (NB) model achieved the highest internal accuracy (82.3%).
    • Logistic Regression (LR), Support Vector Machine (SVM), and NB models demonstrated the highest Area Under the Curve (AUC) of 0.788 internally.
    • External validation confirmed robust generalizability with an AUC of 0.834.
    • SHAP analysis identified estimated glomerular filtration rate (eGFR), serum albumin, C3, serum creatinine, and urinary red blood cell count (URBC) as key predictors.
    • The models effectively captured non-linear patterns, highlighting the predictive value of C3 and URBC.

    Conclusions:

    • Machine learning models can accurately predict prognosis in diabetic kidney disease (DKD).
    • Key predictors such as eGFR, serum albumin, C3, serum creatinine, and URBC provide granular insights for personalized patient management.
    • The study demonstrates the potential of ML in enhancing prognostic assessment for DKD, offering valuable tools for clinicians.