An AI model developed by Johns Hopkins significantly outperforms current risk scores in predicting post-surgical complications using routine ECG data.
Key Details
- 1Johns Hopkins researchers developed AI models to analyze pre-surgical electrocardiograms (ECG).
- 2Models were trained using data from 37,000 surgery patients at Beth Israel Deaconess Medical Center.
- 3The best-performing fusion model predicted serious post-surgical complications with 85% accuracy.
- 4Current clinical risk scoring methods are only about 60% accurate.
- 5The work was federally funded and results were published in the British Journal of Anaesthesia.
- 6The AI can identify previously undetectable signals in ECGs relevant to surgical risk.
Why It Matters
This work offers a major step forward in pre-surgical risk assessment, leveraging an inexpensive, routine test to more accurately identify patients at high risk of serious complications. It could directly impact patient management and surgical planning, highlighting the growing value of AI in medical imaging diagnostics.
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.