A two-stage deep learning pipeline for automated detection and localization of endotracheal tubes on pediatric chest radiographs: a pilot study.
Authors
Affiliations (12)
Affiliations (12)
- Imaging Research Center, Cincinnati Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH, 45229, USA. [email protected].
- Department of Radiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA. [email protected].
- Artificial Intelligence Imaging Research Center, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA. [email protected].
- Department of Radiology, University of Cincinnati College of Medicine, Cincinnati, OH, USA. [email protected].
- Imaging Research Center, Cincinnati Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH, 45229, USA.
- Department of Radiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA.
- Artificial Intelligence Imaging Research Center, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA.
- Department of Radiology, University of Cincinnati College of Medicine, Cincinnati, OH, USA.
- Division of Biostatistics and Epidemiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA.
- Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USA.
- Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA.
- Computer Science, Biomedical Engineering, Biomedical Informatics, University of Cincinnati, Cincinnati, OH, USA.
Abstract
Endotracheal tubes (ETTs) are critical life-support devices for mechanically ventilated pediatric patients, yet automated ETT assessment on pediatric chest radiographs (CXRs) remains limited. To develop and evaluate a two-stage deep learning pipeline for automated detection and localization of ETTs on pediatric CXRs. This retrospective study included 1,000 pediatric CXRs (476 ETT-positive, 524 ETT-negative) acquired in 2021 at a single institution. ETT segmentation masks and distal tip coordinates were annotated by trained analysts and verified by pediatric radiologists. A two-stage pipeline consisting of a ResNet classification model followed by a U-Net segmentation model was developed for ETT detection and localization. Performance was evaluated on a held-out test set using multiple metrics, including the area under the receiver operating characteristic curve (AUROC) and the mean absolute error (MAE), with 95% confidence intervals (CI). Inter-observer variability was assessed as a reference for localization performance. Inter-observer variability for ETT tip localization was 2.01 mm MAE on the held-out test set. The pipeline achieved an AUROC of 0.994 (95% CI 0.986, 1.000) for ETT detection. For localization, the pipeline achieved a MAE of 6.59 mm (95% CI 5.19, 8.21 mm). Incorporating the classification stage substantially reduced false-positive segmentations from 17 to 3 among ETT-negative CXRs compared with the standalone segmentation model. A two-stage deep learning pipeline demonstrated high performance for automated ETT detection and promising performance for tip localization on pediatric CXRs in this single-center pilot study. Further evaluation in larger and external pediatric cohorts is needed to assess generalizability.