Researchers developed AI models using EHR and patient-reported outcomes to identify cancer survivors at high risk for emergency visits and worsening symptoms.
Key Details
- 1Study by Sylvester Comprehensive Cancer Center (University of Miami) published in JCO Clinical Cancer Informatics (May 2026).
- 2AI models analyzed data from over 25,000 cancer survivors followed for three years.
- 3Combined EHR and PRO data to predict unplanned healthcare use (ED visits, hospitalizations) and elevated symptom burden.
- 4Recent clinical activity helped predict acute events, while long-term PRO trends improved symptom prediction.
- 5Adding PROs nearly doubled symptom forecasting accuracy compared to using clinical data alone.
- 6Top 10% of risk-flagged patients accounted for half of downstream healthcare events and symptom episodes.
Why It Matters
Integrating AI-based risk stratification into survivorship care could enable earlier, more targeted interventions for cancer survivors. This demonstrates the evolving potential of medical AI to support proactive patient management across radiology-adjacent fields.
Source
EurekAlert
Related News
•EurekAlert
AI-Enhanced Imaging Breaks Barriers in Complex Media for Biomedical Use
Researchers developed a physics-based machine learning imaging system that enhances visibility through complex media, holding promise for biomedical imaging.
•EurekAlert
AI Model Accelerates Accurate Dose Prediction for CyberKnife Radiotherapy
A physics- and spatially informed diffusion model accelerates Monte Carlo–level dose estimation for CyberKnife radiotherapy using planning CT and anatomical data.
•EurekAlert
New G-AUDIT Tool Detects Hidden Bias in Medical AI Training Data
Johns Hopkins and FDA researchers unveil G-AUDIT, a tool to identify hidden biases in medical AI datasets before model training.