Deep learning for cardiac CT segmentation for congenital heart disease: A systematic review

Sinling Tiffany Yu, Zahra Amini, Shereen Fouad, Jan Novak, Antonio Fratini, Francesco Bardozzo · 2026

Read Deep learning for cardiac CT segmentation for congenital heart disease: A systematic review by Sinling Tiffany Yu, Zahra Amini, Shereen Fouad, Jan Novak, Antonio Fratini, Francesco Bardozzo (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13623130.txt.

Description

Sinling Tiffany Yu and colleagues review deep learning tools that outline the heart on CT scans of patients with congenital heart disease, meaning heart defects present from birth. Drawing these outlines, called segmentation, helps plan surgery, yet each defect is unique and the anatomy is small. No review had gathered this young field before.

Searching six databases for work from 2004 to 2025 left fourteen studies. The UNet design and its variants dominated, and the best score for a single structure came from a hybrid model with a ResNet backbone outlining the aorta. Reporting was strong on training details and model design. Weak spots were plain: only two studies tested their model on outside data, and none explained how they chose sample sizes or handled missing data. Small vessels consistently scored worse than the whole heart. The authors call for bigger shared data sets and standard reporting.

Catalog record

CREATORSinling Tiffany Yu, Zahra Amini, Shereen Fouad, Jan Novak, Antonio Fratini, Francesco Bardozzo
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE35 KB
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13623130

Browse the full library · About the Tower of Babel