AI forecasts risks in ICUs
A new study has shown that artificial intelligence models can achieve over 98% accuracy in predicting the risk of death among intensive care unit (ICU) patients, helping to identify high-risk cases and enable early intervention. Researchers from various Australian universities tested machine learning algorithms such as the Extra Trees model and the Gradient Boosting model, which demonstrated superiority over traditional assessment tools. They identified key factors—including high blood pressure, tumors, and cardiovascular diseases—as primary indicators of elevated risk. The results suggest that combining scientific explanations with AI models enhances understanding of their outcomes and supports proactive care, with plans to test these models on a broader scale across different healthcare settings.
Notice: This Is an AI-Generated Summary
Comments (0)