Detection of pneumoperitoneum in the abdominal radiograph images using artificial neural networks

Mimi Kim; Jong Soo Kim; Changhwan Lee; Bo-Kyeong Kang

doi:10.1016/j.ejro.2020.100316

Detection of pneumoperitoneum in the abdominal radiograph images using artificial neural networks

Eur J Radiol Open. 2020 Dec 21:8:100316. doi: 10.1016/j.ejro.2020.100316. eCollection 2021.

Authors

Mimi Kim¹, Jong Soo Kim², Changhwan Lee³, Bo-Kyeong Kang¹

Affiliations

¹ Department of Radiology, Hanyang University Seoul Hospital, Seoul, Republic of Korea.
² Institute for Software Convergence, Hanyang University, Seoul, Republic of Korea.
³ Department of Biomedical Engineering, Hanyang University, Seoul, Republic of Korea.

Abstract

Background/purpose: The purpose of this study was to assess the diagnostic performance of artificial neural networks (ANNs) to detect pneumoperitoneum in abdominal radiographs for the first time.

Materials and methods: This approach applied a novel deep-learning algorithm, a simple ANN training process without employing a convolution neural network (CNN), and also used a widely utilized deep-learning method, ResNet-50, for comparison.

Results: By applying ResNet-50 to abdominal radiographs, we obtained an area under the ROC curve (AUC) of 0.916 and an accuracy of 85.0 % with a sensitivity of 85.7 % and a predictive value of the negative tests (NPV) of 91.7 %. Compared with the most commonly applied deep-learning methods such as a CNN, our novel approach used extremely small ANN structures and a simple ANN training process. The diagnostic performance of our approach, with a sensitivity of 88.6 % and NPV of 91.3 %, was compared decently with that of ResNet-50.

Conclusions: The results of this study showed that ANN-based computer-assisted diagnostics can be used to accurately detect pneumoperitoneum in abdominal radiographs, reduce the time delay in diagnosing urgent diseases such as pneumoperitoneum, and increase the effectiveness of clinical practice and patient care.

Keywords: Abdominal image; Artificial neural network; Deep learning; Pneumoperitoneum.