Automatic detection of blood content in capsule endoscopy images based on a deep convolutional neural network

Abstract

Background and AimDetecting blood content in the gastrointestinal tract is one of the crucial applications of capsule endoscopy (CE). The suspected blood indicator (SBI) is a conventional tool used to automatically tag images depicting possible bleeding in the reading system. We aim to develop a deep learning‐based system to detect blood content in images and compare its performance with that of the SBI.MethodsWe trained a deep convolutional neural network (CNN) system, using 27 847 CE images (6503 images depicting blood content from 29 patients and 21 344 images of normal mucosa from 12 patients). We assessed its performance by calculating the area under the receiver operating characteristic curve (ROC‐AUC) and its sensitivity, specificity, and accuracy, using an independent test set of 10 208 small‐bowel images (208 images depicting blood content and 10 000 images of normal mucosa). The performance of the CNN was compared with that of the SBI, in individual image analysis, using the same test set.ResultsThe AUC for the detection of blood content was 0.9998. The sensitivity, specificity, and accuracy of the CNN were 96.63%, 99.96%, and 99.89%, respectively, at a cut‐off value of 0.5 for the probability score, which were significantly higher than those of the SBI (76.92%, 99.82%, and 99.35%, respectively). The trained CNN required 250 s to evaluate 10 208 test images.ConclusionsWe developed and tested the CNN‐based detection system for blood content in CE images. This system has the potential to outperform the SBI system, and the patient‐level analyses on larger studies are required.

Artificial intelligence in small bowel capsule endoscopy ‐ current status, challenges and future promise

“…It outperformed earlier classifiers with sensitivity, specificity, and accuracy more than 98% and displayed a segmentation of bleeding zones within abnormal frames. Furthermore, Aoki et al 13 developed a less supervised neural network-based algorithm based on PillCam SBCE still images. This AI solution was trained on 6503 images with blood and 21 334 without blood and then assessed on 208 images with blood and 10 000 without blood.…”

The Future of Capsule Endoscopy: The Role of Artificial Intelligence and Other Technical Advancements

“…In 2019, Aoki et al reported a CNN system for detecting blood content, and compared its performance with that of the suspected blood indicator (SBI), which automatically tags images with suspicious hemorrhages in the reading system. 30 The dataset consisted of 27,847 total CE images, including a training dataset of 6,503 images depicting blood content from 29 patients and a validation dataset of 10,208 images with 208 images depicting blood content. This CNN system outperformed the conventional SBI in terms of sensitivity (96.6% vs. 76.9%), specificity (99.9% vs. 99.8%), and accuracy (99.9% vs. 99.3%).…”

“…The AUROC for detecting angioectasia was 0.998, and the sensitivity and specificity were 98.8% and 98.4%, respectively. In 2019, Aoki et al reported a CNN system for detecting blood content, and compared its performance with that of the suspected blood indicator (SBI), which automatically tags images with suspicious hemorrhages in the reading system [ 30 ]. The dataset consisted of 27,847 total CE images, including a training dataset of 6,503 images depicting blood content from 29 patients and a validation dataset of 10,208 images with 208 images depicting blood content.…”

A New Active Locomotion Capsule Endoscopy under Magnetic Control and Automated Reading Program

“…When CNN-based interpretations were used for the detection of angioectasia, the most common SB vascular lesion, it demonstrated excellent sensitivity and specificity close to 100% [ 45 , 46 ]. In addition, CNN-based interpretations showed an accuracy of 90.8% in the diagnosis of SB erosions and ulcers [ 47 ] and showed a higher diagnostic accuracy in blood contents (sensitivity, 96.63%; specificity, 99.96%) compared to conventional suspected blood indicators [ 48 ]. In a large-scale study in China, a new CNN interpretation model based on 113,426,569 images was developed, which showed a higher sensitivity (77.9%–99.9%) and lesion detection rate (54.6%–70.9%) compared to conventional CE interpretations for detecting various SB lesions, such as inflammation, ulcers, and polyps.…”