Diagnostic Accuracy of AI in Prediction and Assessment of Compromised Free Flaps: Systematic Review and Meta-Analysis.
Kuan-Chen Huang, Melanie J Wang, Yu-Ying Chu et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis evaluated the diagnostic accuracy of AI in monitoring postoperative free flaps for vascular compromise. In 18 studies, AI showed an overall sensitivity of 0.83 and specificity of 0.87. Image-based models outperformed non-image models, achieving a sensitivity of 0.92 and specificity of 0.95, indicating strong potential for clinical application.
Key findings
- Overall sensitivity of AI for flap compromise detection was 0.83 (95% CI 0.70-0.91).
- Image-based AI models had a sensitivity of 0.92 (95% CI 0.81-0.97) and specificity of 0.95 (95% CI 0.86-0.98).
- Pooled sensitivity for vascular compromise detection was 0.92 (95% CI 0.81-0.97) and specificity was 0.91 (95% CI 0.79-0.97).
- The certainty of evidence was rated as low due to methodological concerns and substantial heterogeneity.
Why it matters
The findings suggest that AI can enhance the monitoring of free flaps, potentially improving early detection of vascular compromise. This could lead to better patient outcomes and reduced flap failure rates, making it a valuable tool in surgical practice.
What to keep in mind
The study identified methodological concerns in several included studies and rated the certainty of evidence as low.