DL Development, Validation for Abdominal CT Organ Segmentation in Pediatrics

Transfer learning (TL) models trained on heterogeneous public datasets and fine-tuned using institutional pediatric data outperformed internal native training (NT) models and TotalSegmentator (TS) across internal and external pediatric test data, according to a study published in the American Journal of Roentgenology (AJR).

“The best-performing organ segmentation model, based on internal and external validation, is now available as an open-source application for further research and potential clinical deployment,” noted first author Elanchezhian Somasundaram, PhD, from the radiology department at Cincinnati Children’s Hospital Medical Center.

Somasundaram et al. developed and validated deep-learning models for liver, spleen, and pancreas segmentation using 1,731 CT examinations (1,504 training; 221 testing), derived from three internal institutional pediatric (age ≤ 18) datasets (n = 483) and three public datasets comprising pediatric and adult examinations with various pathologies (n = 1,248). Three deep-learning model architectures (SegResNet, DynUNet, SwinUNETR) from the Medical Open Network for AI (MONAI) framework underwent training using NT, relying solely on institutional datasets, and TL, incorporating pre-training on public datasets. For comparison, TS, a publicly available segmentation model, was applied to test data without further training.

Ultimately, the AJR authors’ optimal model outperformed models trained using only internal pediatric data, as well as a publicly available model. Acknowledging that segmentation performance was better in liver and spleen than in pancreas, “the selected model may be used for various volumetry applications in pediatric imaging,” Somasundaram et al. concluded.