Point-Based Prediction Model for Bladder Cancer Risk in Diabetes: A Random Survival Forest-Guided Approach
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.A new scoring system identifies bladder cancer risk in diabetes patients. Renal dysfunction, indicated by serum creatinine levels, is a key predictor, enabling personalized risk assessment for early detection.
Area Of Science
- Oncology
- Endocrinology
- Data Science
Background
- Diabetes mellitus is linked to increased cancer risk, including bladder cancer.
- Existing bladder cancer risk prediction models for diabetes patients are limited.
- Machine learning approaches can aid in developing novel risk prediction tools.
Purpose Of The Study
- To develop a machine learning-guided scoring system for bladder cancer risk prediction in diabetes patients.
- To identify key predictors of bladder cancer in this population.
- To facilitate individualized risk assessment in routine clinical settings.
Main Methods
- Retrospective cohort study using electronic health records from Hong Kong (2010-2019).
- Random survival forest for variable selection and Cox regression for weight assignment.
- Development of a time-to-event scoring system.
Main Results
- 382,770 diabetes patients were analyzed; 644 developed bladder cancer.
- Key predictors identified: age, serum creatinine, sex, and smoking.
- Serum creatinine ≥94 µmol/L was associated with increased bladder cancer risk; 2- and 5-year AUCs were 0.88 and 0.86.
Conclusions
- Renal dysfunction is a potential bladder cancer predictor in diabetes patients.
- The developed scoring system offers a tool for individualized bladder cancer risk prediction.
- This approach may improve early detection and management strategies for at-risk individuals.
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
