A Lesion-Based Convolutional Neural Network Improves Endoscopic Detection and Depth Prediction of Early Gastric Cancer

Abstract

In early gastric cancer (EGC), tumor invasion depth is an important factor for determining the treatment method. However, as endoscopic ultrasonography has limitations when measuring the exact depth in a clinical setting as endoscopists often depend on gross findings and personal experience. The present study aimed to develop a model optimized for EGC detection and depth prediction, and we investigated factors affecting artificial intelligence (AI) diagnosis. We employed a visual geometry group(VGG)-16 model for the classification of endoscopic images as EGC (T1a or T1b) or non-EGC. To induce the model to activate EGC regions during training, we proposed a novel loss function that simultaneously measured classification and localization errors. We experimented with 11,539 endoscopic images (896 T1a-EGC, 809 T1b-EGC, and 9834 non-EGC). The areas under the curves of receiver operating characteristic curves for EGC detection and depth prediction were 0.981 and 0.851, respectively. Among the factors affecting AI prediction of tumor depth, only histologic differentiation was significantly associated, where undifferentiated-type histology exhibited a lower AI accuracy. Thus, the lesion-based model is an appropriate training method for AI in EGC. However, further improvements and validation are required, especially for undifferentiated-type histology.

Convolutional Neural Network Technology in Endoscopic Imaging: Artificial Intelligence for Endoscopy

“…Transfer learning for low-level CNN features from various non-medical source tasks using deep CNN representations showed good performance for the automatic detection and classification of colorectal polyps. 44,46,47 The proposed diagnostic model, which minimized the time-consuming pre-processing, outperformed the previous state-of-the-art methods.…”

Lesion-Based Convolutional Neural Network in Diagnosis of Early Gastric Cancer

“…Yoon et al named this process, "lesion-based CNN". 36 They reported that it showed better performance than other CNNs in the detection of EGC and depth prediction. The Grad-CAM loss was applied to the existing cross-entropy loss to reduce localization errors.…”

Application of artificial intelligence in gastrointestinal endoscopy

“…Smart AI‐assisted systems integrating detection and diagnosis will become a new focus of research . Mori et al worked on an automated colonoscopic observation that is hoped to be able to detect and characterize diminutive polyps simultaneously.…”

“…As assistant observers or clinical decision‐makers, AI‐aided systems undoubtedly have a promising future in increasing the effective training of junior endoscopists . Furthermore, it has emerged that, given the advantages of computer vision, the depth of tumor invasion can be predicted …”

“…These limitations involve aspects such as the datasets used, trial design, and outcome indicators. First, the researchers tend to collect high‐quality endoscopic images to construct training sets and exclude common low‐quality and unanalyzable images (eg, obscure images, those containing mucus, stained images, or those with partial views of lesions), which may lead to poor generalization in models in clinical practice . In this situation, the accuracy of CAD might be exaggerated, as its good performance when using static, high‐quality images does not guarantee that recognition will be successful when using dynamic videos.…”