Artificial intelligence-aided colonoscopy: Recent developments and future perspectives

Abstract

Artificial intelligence (AI) systems, especially after the successful application of Convolutional Neural Networks, are revolutionizing modern medicine. Gastrointestinal Endoscopy has shown to be a fertile terrain for the development of AI systems aiming to aid endoscopists in various aspects of their daily activity. Lesion detection can be one of the two main aspects in which AI can increase diagnostic yield and abilities of endoscopists. In colonoscopy, it is well known that a substantial rate of missed neoplasia is still present, representing the major cause of interval cancer. In addition, an extremely high variability in adenoma detection rate, the main key quality indicator in colonoscopy, has been extensively reported. The other domain in which AI is believed to have a considerable impact on everyday clinical practice is lesion characterization and aid in “optical diagnosis”. By predicting in vivo histology, such pathology costs may be averted by the implementation of two separate but synergistic strategies, namely the “leave-in-situ” strategy for < 5 mm hyperplastic lesions in the rectosigmoid tract, and “resect and discard” for the other diminutive lesions. In this opinion review we present current available evidence regarding the role of AI in improving lesions’ detection and characterization during colonoscopy.

Artificial Intelligence-Aided Endoscopy and Colorectal Cancer Screening

“…Secondly, because Computer-Aided Detection systems are based on deep learning architectures in which there is no human control on the final algorithm, their outcomes incorporate some unpredictability, and the endoscopist may be presented with false positive (FP) images which should be never considered as suspicious [ 83 , 84 , 85 ]. Such FPs may potentially hamper the efficiency of Computer-Aided Detection-assisted colonoscopy due to increased withdrawal time, unneeded polypectomies, and higher endoscopist’s fatigue and distraction [ 86 , 87 , 88 ]. However, despite a relevant number of FPs per colonoscopy (27.3 ± 13.1), only a minority of them (5.7%) required additional exploration, resulting in a negligible 1% increase in the total withdrawal time [ 89 ] in a post hoc analysis of an RCT [ 30 ], and confirmed by a following comparative study [ 90 ] ( Figure 4 ).…”

Application of Artificial Intelligence in Gastroenterology: A Review on Improvement in Diagnosis, Treatment, and Patient Monitoring

“…AI has been used to aid in diagnosis, assess the severity of intestinal damage, and identify patients who may benefit from further diagnostic testing. Machine learning algorithms have been developed to analyze endoscopic images and biopsies of the small intestine to identify and classify the severity of villous atrophy, which is a hallmark of coeliac disease 8 . AI algorithms can help pathologists identify precancerous and cancerous cells, as well as distinguish between inflammatory bowel disease and other conditions that may present with similar symptoms.…”

A Review of Convolutional Neural Network Development in Computer Vision

“…Average pooling can be regarded as a structural regularization, which can improve the consistency between feature surfaces and categories [36][37][38]. There are no parameters that need to be optimized in the globally averaged sampling layer, so overfitting can be avoided.…”