July 30, 2025Abdominal RadiologyOpen Access

From promise to practice: a scoping review of AI applications in abdominal radiology

View Full Paper

Ask AI

Bookmark

AI is transforming abdominal radiology, with a focus on segmentation, detection, and classification across 432 studies.
Segmentation models like U-Net excelled in liver imaging, achieving Dice coefficients between 0.65 and 0.90.
Classification models showed sensitivities from 52 to 100% and specificities from 40.7 to 99% for diagnostic labeling.
Major gaps include limited real-world validation and the need for standardized benchmarks in AI-driven workflows.

Discussion

No takes yet. Share an insight, caveat, or question.

Scoping review maps AI applications in CT imaging and workflow optimization, indicating key trends and gaps.

Fotis et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark