Multimodal deep learning model for AI-based functional prognostic risk stratification in patients undergoing radical nephrectomy

2026-05-28 · Source: Machine learning : nature.com subject feeds · Field: Health & Wellbeing — Health & Medical Research, Medical Devices & Health Technology, Artificial Intelligence & Machine Learning · Depth: Expert, medium

Summary

A multimodal deep learning model has been developed to predict rapid glomerular filtration rate (GFR) decline after radical nephrectomy (RN) in patients with complex renal cell carcinoma (RCC). The model was trained and retrospectively analyzed using contrast-enhanced computed tomography images and clinical data from 1621 patients across multiple centers. It achieved an area under the curve (AUC) of 0.788–0.873 in external test sets. This AI-based tool stratifies patients into high- and low-risk groups for chronic kidney disease progression, demonstrating potential to assist urologists in treatment decisions, especially when technically challenging partial nephrectomy (PN) is a feasible alternative to RN.

Key takeaway

For urologists managing complex renal cell carcinoma patients, this model offers a data-driven tool to predict the risk of rapid GFR decline post-radical nephrectomy. Your ability to preoperatively identify high-risk patients could influence surgical planning, potentially favoring a technically demanding partial nephrectomy to preserve renal function. Consider integrating such AI-based prognostic tools into your clinical workflow for enhanced decision support.

Key insights

A multimodal deep learning model predicts post-surgical kidney function decline in complex renal cell carcinoma patients.

Principles

• Preoperative risk prediction can guide surgical choice.
• Multimodal data improves prognostic stratification.

Method

A multimodal deep learning model integrates contrast-enhanced CT images and clinical data from 1621 patients to predict rapid GFR decline after radical nephrectomy.

In practice

• Stratify patients into high- and low-risk groups.
• Inform decision-making between PN and RN.

Topics

• Multimodal Deep Learning
• Renal Cell Carcinoma
• Radical Nephrectomy
• Glomerular Filtration Rate
• Prognostic Risk Stratification
• Urologic Oncology

Best for: AI Scientist, Research Scientist, Data Scientist

Related on AIssential

• 🔬 Training Transformers to solve 95% failure rate of Cancer Trials — Ron Alfa & Daniel Bear, Noetik — Improving cancer drug success requires advanced AI models for precise patient selection based on multimodal biological data.
• A deep learning–based digital biopsy for predicting early recurrence in gastric cancer — The RSA model combines deep learning with clinical data to predict gastric cancer recurrence.
• I Gave a Model Eyes and a Genome. Here’s What It Learned About Brain Cancer. — Cross-attention fusion of pathology and genomics significantly improves glioblastoma survival prediction over unimodal approaches.
• Multi-modal AI for comprehensive breast cancer prognostication — A multi-modal AI test significantly improves breast cancer prognostication by integrating pathology and clinical data.
• A deep learning framework for glomeruli segmentation with boundary attention — A U-Net-based model with a boundary attention decoder improves glomeruli segmentation in kidney tissue.
• Validating a Deep Learning Algorithm to Identify Patients with Glaucoma using Systemic Electronic Health Records — An EHR-only deep learning model can effectively pre-screen for glaucoma without specialized imaging.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.