Computer-Aided Diagnosis of Pneumoperitoneum on Neonatal Abdominal Radiographs
Neonatology · 2025
DOI: 10.1159/000549186 ↗Kurz gefasst
Multizentrische Studie in der Zeitschrift Neonatology, 2025. Verfasst von Yohei Sanmoto und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pediatric Surgery, University of Tsukuba Hospital, Tsukuba, Japan. Hauptthemen laut MEDLINE: Pneumoperitoneum, Radiography, Abdominal, Intestinal Perforation, Diagnosis, Computer-Assisted. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 23 Literaturangaben.
Abstract (Originalsprache)
Introduction
Neonatal gastrointestinal perforation is a life-threatening condition that requires timely and accurate diagnosis. However, interpreting abdominal radiographs in this population is often challenging. In this study, we aimed to develop a deep convolutional neural network (DCNN) model to segment pneumoperitoneum on neonatal abdominal radiographs and to evaluate its potential to assist in detecting neonatal gastrointestinal perforation.
Methods
This multicenter retrospective study included 1,187 abdominal radiographs (181 perforation and 1,006 control images) from neonates with gastrointestinal perforation and controls. Pneumoperitoneum regions were annotated by experienced clinicians. The dataset was randomly divided into training (n = 830), validation (n = 118), and test (n = 239) sets. A DeepLabV3+ model with ResNet50 backbone was fine-tuned for pixel-level segmentation. A single pixel-based threshold, derived from ROC analysis, was used to classify gastrointestinal perforation, with diagnostic performance subsequently compared to that of clinicians.
Results
The DCNN model achieved a median Dice similarity coefficient of 0.81 on the test dataset, indicating strong overlap between predicted and actual pneumoperitoneum regions. Furthermore, segmentation performance was positively correlated with pneumoperitoneum volume (Spearman ρ = 0.83, p < 0.001). Classification using the pixel-based cut-off demonstrated excellent diagnostic accuracy (AUC, 0.999; sensitivity, 100%; specificity, 98.5%), comparable to experienced clinicians.
Conclusion
The DCNN model demonstrated robust segmentation and classification performance, highlighting its potential as a clinical decision support tool for early detection of gastrointestinal perforation in neonates. Future studies should validate the model’s generalizability and assess its integration into clinical practice.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Pediatric Surgery, University of Tsukuba Hospital, Tsukuba, Japan.
- Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, Japan.
- Graduate School of Computer Science and Engineering, The University of Aizu, Aizu, Japan.
- Information Center, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China.
- Department of Radiology, Saitama Children's Medical Center, Saitama, Japan.
- Department of Pediatric Surgery, Saitama Children's Medical Center, Saitama, Japan.
Zitieren
APA
Sanmoto, Y., Zhang, R., Peng, B., Hosokawa, T., Kondo, Y., Inoue, M., Miyazono, Y., Zhu, X., & Masumoto, K. (2025). Computer-Aided Diagnosis of Pneumoperitoneum on Neonatal Abdominal Radiographs. Neonatology. https://doi.org/10.1159/000549186
Vancouver
Sanmoto Y, Zhang R, Peng B, Hosokawa T, Kondo Y, Inoue M, et al. Computer-Aided Diagnosis of Pneumoperitoneum on Neonatal Abdominal Radiographs. Neonatology. 2025. doi:10.1159/000549186
BibTeX
@article{Sanmoto2025,
author = {Sanmoto, Yohei and Zhang, Ruiyao and Peng, Boyuan and Hosokawa, Takahiro and Kondo, Yasuhiro and Inoue, Mikihiro and Miyazono, Yayoi and Zhu, Xin and Masumoto, Kouji},
title = {Computer-Aided Diagnosis of Pneumoperitoneum on Neonatal Abdominal Radiographs},
journal = {Neonatology},
year = {2025},
doi = {10.1159/000549186}
}
EUVIMED