VEELA: A Clinically-Constrained Benchmark for Liver Vessel Segmentation in Computed Tomography Angiography
Researchers have introduced VEELA, a new benchmark dataset designed for segmenting liver vessels in computed tomography angiography (CTA) scans. This dataset, derived from 40 CTA scans and curated with multi-expert consensus, addresses the challenges of complex vascular topology and limited visibility in existing benchmarks. VEELA aims to facilitate reproducible research by providing a standardized framework and complementary evaluation metrics to develop more robust vessel segmentation methods. AI
IMPACT Provides a new, clinically-constrained dataset and evaluation framework to advance AI-driven liver vessel segmentation in medical imaging.