AI-driven text simplification significantly improves cancer patients' comprehension of CT scan reports.
Key Details
- 1Prospective controlled trial at TUM included 200 cancer patients undergoing CT imaging.
- 2Half received original reports; half got AI-simplified versions using a local large language model.
- 3Patients with simplified reports had reduced reading time (from 7 to 2 minutes).
- 4Comprehension ratings: 81% found simplified reports easy to read versus 17% with originals; 80% easier to understand (vs 9%).
- 5Incidence of AI factual errors was 6%; omissions 7%; additions 3%, but all reports were reviewed and corrected by radiologists.
- 6Study published in 'Radiology' (DOI: 10.1148/radiol.251844).
Why It Matters
AI simplification can greatly enhance patient empowerment and health literacy by making complex radiology findings accessible. However, human oversight remains essential to prevent misinformation, highlighting the need for secure and clinically integrated AI solutions.
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.