Physician-guided AI framework improves delirium detection in hospitals

Researchers have developed a user-centered interactive machine learning (UC-iML) framework to assist physicians in identifying delirium in hospitalized patients. This framework integrates various patient data, including administrative variables, lab results, and medications, with physician-guided feature refinement and interpretable modeling using SHAP explanations. Tested on data from six Toronto hospitals, the UC-iML approach demonstrated improved discrimination and temporal robustness compared to automated and baseline methods, highlighting its potential as a practical tool for clinical delirium modeling. AI

IMPACT This framework could enhance clinical decision support systems, leading to earlier and more accurate diagnosis of delirium in hospital settings.

RANK_REASON Academic paper detailing a new machine learning framework. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Delirium
• General Medicine Inpatient Initiative (GEMINI)
• Machine Learning
• Physician

• paper
• product
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →