Upper gastrointestinal anatomy detection with multi-task convolutional neural networks.

Journal: Healthcare technology letters
Published Date:

Abstract

Esophagogastroduodenoscopy (EGD) has been widely applied for gastrointestinal (GI) examinations. However, there is a lack of mature technology to evaluate the quality of the EGD inspection process. In this Letter, the authors design a multi-task anatomy detection convolutional neural network (MT-AD-CNN) to evaluate the EGD inspection quality by combining the detection task of the upper digestive tract with ten anatomical structures and the classification task of informative video frames. The authors' model is able to eliminate non-informative frames of the gastroscopic videos and detect the anatomies in real time. Specifically, a sub-branch is added to the detection network to classify NBI images, informative and non-informative images. By doing so, the detected box will be only displayed on the informative frames, which can reduce the false-positive rate. They can determine the video frames on which each anatomical location is effectively examined, so that they can analyse the diagnosis quality. Their method reaches the performance of 93.74% mean average precision for the detection task and 98.77% accuracy for the classification task. Their model can reflect the detailed circumstance of the gastroscopy examination process, which shows application potential in improving the quality of examinations.

Authors

  • Zhang Xu
    Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou 310027, People's Republic of China.
  • Yu Tao
    Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou 310027, People's Republic of China.
  • Zheng Wenfang
    Department of Gastroenterology, Sir Run Run Shaw Hospital, Medical School, Zhejiang University, Hangzhou, 310016, People's Republic of China.
  • Lin Ne
    Department of Gastroenterology, Sir Run Run Shaw Hospital, Medical School, Zhejiang University, Hangzhou, 310016, People's Republic of China.
  • Huang Zhengxing
    Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou 310027, People's Republic of China.
  • Liu Jiquan
    Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou 310027, People's Republic of China.
  • Hu Weiling
    Department of Gastroenterology, Sir Run Run Shaw Hospital, Medical School, Zhejiang University, Hangzhou, 310016, People's Republic of China.
  • Duan Huilong
    Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou 310027, People's Republic of China.
  • Si Jianmin
    Department of Gastroenterology, Sir Run Run Shaw Hospital, Medical School, Zhejiang University, Hangzhou, 310016, People's Republic of China.

Keywords

No keywords available for this article.