人工智能在胶囊内镜中的应用现状与前景

Application status and prospects of artificial intelligence in capsule endoscopy

以深度学习中卷积神经网络模型为代表的人工智能(artificial intelligence,AI)技术为胶囊内镜(capsule endoscopy,CE)提供了一种高效而准确的自动图像识别方法,可辅助临床医生诊断,在多项研究中具有较好的应用效果。同时,人工智能在多病变分类诊断、胶囊内镜定位、内镜质量控制、胶囊内镜教学培训以及多场景应用等方面均具有较大的发展潜力。

Artificial intelligence(AI),especially the deep learning with the use of convolutional neural networks,provides an efficient and accurate method for capsule endoscopy(CE)to automatically recognize images,which can aid endoscopists in the clinical diagnosis and has satisfactory effect in multiple studies.Meanwhile,AI has great potential of development for CE’s applications in many aspects such as multi-lesion classification,capsule localization,quality control and training of CE procedure,and multi-scenario application.