Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster display session

174P - Progression of computer aided diagnosis on gastric cancer


23 Nov 2019


Poster display session


Tumour Site

Gastric Cancer


Yingyan Yu


Annals of Oncology (2019) 30 (suppl_9): ix42-ix67. 10.1093/annonc/mdz422


Y. Yu

Author affiliations

  • Department Of Surgery, Shanghai Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 200025 - Shanghai/CN


Login to get immediate access to this content.

If you do not have an ESMO account, please create one for free.

Abstract 174P


Application of computer and artificial intelligence (AI) has made revolutionary progress in medical fields. Computer-assisted diagnosis (CAD) on pathology is successful in diagnosis, classification, staging and predicting prognosis of cancer.


In this work, we used the technique of convolutional neural networks (CNN) for training of morphological images of gastric cancer, which is a promising application of artificial intelligence technique for medical image analysis, especially for cancer pathology.


We introduced successful works of AI-aided gastric cancer analysis for over 300 cases. AI-assisted pathology technique could help the pathologist in detecting and locating the abnormal area in images and to diagnose and classify benign or malignant tissues. Therefore, AI exactly improves the diagnostic status of gastric cancer, such as in interpretation of medical images and classification between benign and cancerous tissues. The complete workflow of AI diagnosis for gastric cancer is under construction. AI is more useful in releasing pathologists from repeated work, and improving efficiency.


In conclusions, there are still notable technical obstacles before this approach can be used to improve conventional clinical practice. Higher performance needs corrected marking images by experienced pathologist and require ‘training’ algorithms on tremendous data. We stressed the importance of proper algorithms for improving confidence of analytic results that is close to human experts. Additionally, we indicated the gaps in current research and principal resolutions for advancing the field.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors without further recourse to the authors.


Has not received any funding.


All authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.