Key result
Automated PE segmentation fails in ~43% of CTPA exams due to lung disease, anatomy, and scan parameters.
Why the study?
Automated pulmonary embolism segmentation is frequently used but its potential limitations in clinical practice have not been fully analyzed.
Cross-Sectional (n=304)
Automated pulmonary embolism segmentation on CTPA frequently fails due to specific lung diseases, anatomical variations, and scan parameters, which must be addressed to develop robust automated schemes.
Answers are grounded in this paper and cite their sources.
Frequent failures preclude routine clinical adoption of automated PE segmentation; leaves open targeted algorithm improvements for lung diseases and scan variations.
No abstract yet. The journal hasn't published one we can show.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2018) conducted a cross-sectional in Pulmonary embolism (n=304). Automated pulmonary embolism (PE) segmentation was evaluated on Classification of segmentation results into four groups (A, B, C, D) and identification of failure reasons. Automated pulmonary embolism segmentation on 304 CTPA exams resulted in failed segmentation in 43.1% of cases due to specific lung diseases, anatomy varieties, and improper scan parameters.
Synapse has enriched 10 related papers. Consider them for comparative context: