Houston Methodist researchers used unsupervised machine learning to identify distinct risk clusters in patients with bloodstream infections, aiding early intervention.
Key Details
- 1Study analyzed data from over 15,000 patients with bloodstream infection (BSI) using EHR-derived clinical variables.
- 2Unsupervised machine learning clustered patients into three clinically distinct groups based on data from the first 48 hours of BSI diagnosis.
- 3High-risk cluster included older, predominantly male, and transplant patients, with mortality rates up to 60%.
- 4Model converts early clinical data into a risk map for immediate clinical use, potentially accelerating intervention.
- 5Findings to be validated further across external healthcare systems; published in the American Journal of Transplantation.
Why It Matters
Early risk stratification of BSI patients, especially transplant recipients, can prompt timely intervention and improve survival rates. This demonstrates the expanding role of AI and machine learning in data-driven clinical support for high-acuity populations.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.