Gastritis 2 Classification Dataset

Citation Author(s):
Dawei
Gong
Lingling
Yan
Binbin
Gu
Ruili
Zhang
Xinli
Mao
Sailing
He
Submitted by:
Sailing He
Last updated:
Thu, 10/19/2023 - 00:07
DOI:
10.21227/3yqx-d789
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Endoscopic images of patients with AIG, type B gastritis, and CNAG were collected between January 1st, 2019 and March 1st, 2023. All endoscopic images were acquired by Olympus Evis Lucera 260/290 (Tokyo, Japan) and FUJIFILM EC-760RV/M (Tokyo, Japan). This dataset includes 3576 endoscopic images. Poor image quality, images of the pharynx, images of the esophageal region, and images of the duodenal region were excluded. The patient name, age, gender, date, and other irrelevant information for classification in the endoscopic image are cropped out and only the pure endoscopic image of the stomach is retained. The training set which accounts for 70% of the total and the validation set which accounts for 30% of the total are split by randomly shuffling the images.

Instructions: 

You can unzip gastric2cls.zip. We haved finished the split of training and validation set.

Comments

ok

Submitted by Harsha Yalla on Wed, 04/03/2024 - 21:06