Artificial Intelligence-Assisted Liver Ultrasound Training: Image Recognition and Real-Time Feedback System Based on YOLO and DeepSeek.
Acad RadiolAug 7, 2026 (epub)
Ting Liu, Qin Wen, Jiaqi Yang et al.
✦ AI-curated · Sources linked
In 30 seconds
This randomized controlled trial evaluated an AI-assisted liver ultrasound training system using the YOLOv11n algorithm and DeepSeek model among 20 students. The AI system significantly improved training outcomes, with post-training examination scores increasing from 87.9 to 96.2 points. Additionally, it reduced report processing time by 96.3% and feedback latency by 95.3%.
Key findings
- The YOLOv11n model achieved a precision of 0.984 and a recall of 1.00.
- Post-training examination scores improved from 87.9 to 96.2 points.
- Daily report processing time decreased from 486 to 18 minutes.
- Section recognition accuracy increased from 86.0% to 96.0%.
Why it matters
The integration of AI in ultrasound training can enhance educational efficiency and improve learning outcomes, potentially leading to better diagnostic skills among practitioners.