According to the Ministry of Health, Labor and Welfare's "2016 Vital Statistics Overview", the number one cause of death in Japan is malignant neoplasm (cancer) for both men and women.It accounts for 1% of the causes of death, and gastric cancer is the second most common cause of death after lung cancer in men and the fourth most common cause in females.In terms of the number of cancer patients, it ranks first among men and third among females, and the 28.5-year relative survival rate is 2% for males and 4% for females.There is a high possibility that it will be cured by early treatment, and it is important how to preserve the function after surgery.

　In this joint research, AI is made to learn the "metastasis lymph node histology" processed by the HE staining method *, and is it possible for AI to diagnose metastasis at a level equal to or higher than that of a pathologist? Verify.

　If the effectiveness of AI-assisted metastasis diagnosis is high, not only will the burden on pathologists be significantly reduced, but in the future, postoperative QOL (Quality Of Life) will be achieved by reduced surgery that preserves gastric function. Improvement is also expected.

　Professor Hayashi commented on the joint research, "In rapid histological diagnosis, which requires the presence or absence of minute metastases during surgery, it is necessary to make a diagnosis with high accuracy within a limited time. If AI technology can be used to check all HE-stained tissue specimens used by pathologists for final diagnosis in a short period of time, not only will the burden on pathologists be significantly reduced, but also cancer It is expected that "body-friendly surgery" that can improve postoperative QOL with a minimum of excision without sacrificing curability will become widespread in the world. " There is.

* HE staining method is a method of staining cell nuclei and cytoplasm using hematoxylin and eosin, respectively.