Yoon Hong Jin, Kim Seunghyup, Kim Jie-Hyun, Keum Ji-Soo, Oh Sang-Il, Jo Junik, Chun Jaeyoung, Youn Young Hoon, Park Hyojin, Kwon In Gyu, Choi Seung Ho, Noh Sung Hoon
Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul 06273, Korea.
SELVAS AI Inc., Seoul 08594, Korea.
J Clin Med. 2019 Aug 26;8(9):1310. doi: 10.3390/jcm8091310.
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.
在早期胃癌（EGC）中，肿瘤浸润深度是决定治疗方法的重要因素。然而，由于内镜超声在临床环境中测量确切深度时存在局限性，因为内镜医师通常依赖大体所见和个人经验。本研究旨在开发一种针对EGC检测和深度预测进行优化的模型，并且我们调查了影响人工智能（AI）诊断的因素。我们采用视觉几何组（VGG）-16模型对内镜图像进行分类，判断其为EGC（T1a或T1b）还是非EGC。为了在训练期间促使模型激活EGC区域，我们提出了一种同时测量分类和定位误差的新型损失函数。我们用11539幅内镜图像（896例T1a期EGC、809例T1b期EGC和9834例非EGC）进行了实验。用于EGC检测和深度预测的受试者操作特征曲线下面积分别为0.981和0.851。在影响AI预测肿瘤深度的因素中，只有组织学分化显著相关，其中未分化型组织学表现出较低的AI准确性。因此，基于病变的模型是EGC中AI的一种合适训练方法。然而，特别是对于未分化型组织学，还需要进一步改进和验证。
