Abstract
Ancylostomiasis is a fairly common small bowel parasite disease identified by capsule endoscopy (CE) for which a computer-aided clinical detection method has not been established. We sought to develop an artificial intelligence system with a convolutional neural network (CNN) to automatically detect hookworms in CE images. We trained a deep CNN system based on a YOLO-V4 (You Look Only Once-Version4) detector using 11236 CE images of hookworms. We assessed its performance by calculating the area under the receiver operating characteristic curve and its sensitivity, specificity, and accuracy using an independent test set of 10,529 small-bowel images including 531 images of hookworms. The trained CNN system required 403 seconds to evaluate 10,529 test images. The area under the curve for the detection of hookworms was 0.972 (95% confidence interval (CI), 0.967-0.978). The sensitivity, specificity, and accuracy of the CNN system were 92.2%, 91.1%, and 91.2%, respectively, at a probability score cut-off of 0.485. We developed and validated a CNN-based system for detecting hookworms in CE images. By combining this high-accuracy, high-speed, and oversight-preventing system with other CNN systems, we hope it will become an important supplement for detecting intestinal abnormalities in CE images. This trial is registered with ChiCTR2000034546 (a clinical research of artificial-intelligence-aided diagnosis for hookworms in small intestine by capsule endoscope images).
Highlights
Remarkable progression in the investigation and diagnosis of small bowel lesions, such as tumors, ulcerations, enteritis, and parasites, by capsule endoscopy (CE) has been made in recent years [1,2,3,4]
The most common cause of hookworm infection was touching soil containing filariform larvae of hookworms with bare hands, feet, or other parts of the body or consuming food containing filariform larvae of hookworms, but in both datasets, the patients could not provide the relevant information of history of infection due to chronic and occult incidence
We developed a convolutional neural network (CNN)-based system for automatic detection of hookworms in small bowel CE images
Summary
Remarkable progression in the investigation and diagnosis of small bowel lesions, such as tumors, ulcerations, enteritis, and parasites, by capsule endoscopy (CE) has been made in recent years [1,2,3,4]. With the continuous development of the combination of computer software technology and endoscopic diagnosis [5], many computer-aided methods have been formed, and such methods are promising for the detection of many small intestinal abnormalities [6,7,8], such as bleeding [9], erosions [1], ulcerations [1], angioectasias [10], and protruding lesions [11], such as polyps, nodules, epithelial tumors, stromal tumors, and venous structures. In our earlier report on automatic detection software based on the color and morphological features of hookworms [13], the ability of the software to detect hookworms was Gastroenterology Research and Practice poorer than its ability to detect the lesions mentioned above due to algorithm imperfections [13]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
