An artificial intelligence deep learning model for identification of small bowel obstruction on plain abdominal radiographs

Abstract

Objectives: Small bowel obstruction is a common surgical emergency which can lead to bowel necrosis, perforation and death. Plain abdominal X-rays are frequently used as a first-line test but the availability of immediate expert radiological review is variable. The aim was to investigate the feasibility of using a deep learning model for automated identification of small bowel obstruction.

Methods: A total of 990 plain abdominal radiographs were collected, 445 with normal findings and 445 demonstrating small bowel obstruction. The images were labelled using the radiology reports, subsequent CT scans, surgical operation notes and enhanced radiological review. The data were used to develop a predictive model comprising an ensemble of five convolutional neural networks trained using transfer learning.

Results: The performance of the model was excellent with an area under the receiver operator curve (AUC) of 0.961, corresponding to sensitivity and specificity of 91 and 93% respectively.

Conclusion: Deep learning can be used to identify small bowel obstruction on plain radiographs with a high degree of accuracy. A system such as this could be used to alert clinicians to the presence of urgent findings with the potential for expedited clinical review and improved patient outcomes.

Advances in knowledge: This paper describes a novel labelling method using composite clinical follow-up and demonstrates that ensemble models can be used effectively in medical imaging tasks. It also provides evidence that deep learning methods can be used to identify small bowel obstruction with high accuracy.

Conflict of interest statement

Competing interests: Dr Daniel H Kim and Dr G Maskell are providing support to core research projects for the National Consortium of Intelligent Medical Imaging (NCIMI) which currently involves collaboration with the University of Oxford, General Electric and Alliance Medical. Dr G Maskell is currently on the board for NCIMI. Dr Daniel H Kim is on the Data Access Committee for NCIMI. The submitted study received no funding or support from NCIMI or any of its partners; or any other external institution. The remaining authors declare that they have no competing financial interests or personal relationships that could have appeared to influence the work reported in this.

Figures

Figure 1.

Flow diagram illustrating image labelling and network training methods. CNN, convolution neural network; PACS, picture archiving and communication system

Figure 2.

ROC curve for the ensemble model. ROC, receiver operator characteristic.

Figure 3.

Examples of correctly categorised images demonstrating normal appearances (a) and obstruction (b). The ensemble model scores and category predictions are shown in boxes at the base of each image.

Figure 4.

Examples of incorrectly diagnosed images showing false negatives (a) and false positives (b). Ensemble model output scores and class predictions are shown in boxes at the base of each image.