Risk prediction in IgA nephropathy: from conventional models to machine learning, deep learning, and precision nephrology - PubMed
3 months ago
- #IgA nephropathy
- #precision nephrology
- #machine learning
- IgA nephropathy (IgAN) is the most common primary glomerular disease and a major cause of end-stage kidney disease (ESKD).
- Early identification of high-risk patients is crucial due to the disease's clinical heterogeneity.
- Prognostic models help stratify ESKD risk, guide treatment, and optimize intervention timing.
- Traditional models like the International IgA Nephropathy Prediction Tool (IIgAN-PT) are widely used but have limitations in reflecting dynamic disease progression.
- Machine learning (ML) and deep learning (DL) models improve predictive accuracy by integrating multi-omics data and digital pathology.
- These advanced models support dynamic risk tracking and personalized treatment predictions.
- The review discusses the evolution of IgAN prognostic models, their strengths, limitations, and future trends like explainable AI and multimodal prognostication.