An Optimal Artificial Intelligence System for Real-Time Endoscopic Prediction of Invasion Depth in Early Gastric Cancer

• Published in: Cancers Vol. 14; no. 23; p. 6000
• Main Authors: Kim, Jie-Hyun, Oh, Sang-Il, Han, So-Young, Keum, Ji-Soo, Kim, Kyung-Nam, Chun, Jae-Young, Youn, Young-Hoon, Park, Hyojin
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.12.2022; MDPI
• Subjects: Artificial intelligence; Brain; Cancer invasiveness; convolutional neural networks; Datasets; Deep learning; Development and progression; Endoscopy; Gastric cancer; Health aspects; Methods; Neural networks; Performance evaluation; Prognosis; Standard deviation; Stomach cancer; Technology application; video; South Korea; endoscopy; convolutional neural networks; video; gastric cancer; artificial intelligence
• ISSN: 2072-6694; 2072-6694
• DOI: 10.3390/cancers14236000

We previously constructed a VGG-16 based artificial intelligence (AI) model (image classifier [IC]) to predict the invasion depth in early gastric cancer (EGC) using endoscopic static images. However, images cannot capture the spatio-temporal information available during real-time endoscopy-the AI trained on static images could not estimate invasion depth accurately and reliably. Thus, we constructed a video classifier [VC] using videos for real-time depth prediction in EGC. We built a VC by attaching sequential layers to the last convolutional layer of IC 2, using video clips. We computed the standard deviation (SD) of output probabilities for a video clip and the sensitivities in the manner of frame units to observe consistency. The sensitivity, specificity, and accuracy of IC 2 for static images were 82.5%, 82.9%, and 82.7%, respectively. However, for video clips, the sensitivity, specificity, and accuracy of IC 2 were 33.6%, 85.5%, and 56.6%, respectively. The VC performed better analysis of the videos, with a sensitivity of 82.3%, a specificity of 85.8%, and an accuracy of 83.7%. Furthermore, the mean SD was lower for the VC than IC 2 (0.096 vs. 0.289). The AI model developed utilizing videos can predict invasion depth in EGC more precisely and consistently than image-trained models, and is more appropriate for real-world situations.