A deep neural network improves endoscopic detection of early gastric cancer without blind spots

Endoscopy ◽  
2019 ◽  
Vol 51 (06) ◽  
pp. 522-531 ◽  
Author(s):  
Lianlian Wu ◽  
Wei Zhou ◽  
Xinyue Wan ◽  
Jun Zhang ◽  
Lei Shen ◽  
...  
Keyword(s):  
Neural Network ◽  
Gastric Cancer ◽  
Early Gastric Cancer ◽  
Predictive Value ◽  
The Third ◽  
Visual Tasks ◽  
Deep Convolution Neural Network ◽  
Different Parts ◽  
Better Than ◽  
Blind Spots

Abstract Background Gastric cancer is the third most lethal malignancy worldwide. A novel deep convolution neural network (DCNN) to perform visual tasks has been recently developed. The aim of this study was to build a system using the DCNN to detect early gastric cancer (EGC) without blind spots during esophagogastroduodenoscopy (EGD). Methods 3170 gastric cancer and 5981 benign images were collected to train the DCNN to detect EGC. A total of 24549 images from different parts of stomach were collected to train the DCNN to monitor blind spots. Class activation maps were developed to automatically cover suspicious cancerous regions. A grid model for the stomach was used to indicate the existence of blind spots in unprocessed EGD videos. Results The DCNN identified EGC from non-malignancy with an accuracy of 92.5 %, a sensitivity of 94.0 %, a specificity of 91.0 %, a positive predictive value of 91.3 %, and a negative predictive value of 93.8 %, outperforming all levels of endoscopists. In the task of classifying gastric locations into 10 or 26 parts, the DCNN achieved an accuracy of 90 % or 65.9 %, on a par with the performance of experts. In real-time unprocessed EGD videos, the DCNN achieved automated performance for detecting EGC and monitoring blind spots. Conclusions We developed a system based on a DCNN to accurately detect EGC and recognize gastric locations better than endoscopists, and proactively track suspicious cancerous lesions and monitor blind spots during EGD.

scholarly journals Predictive value of risk score using Kyoto classification of gastritis a few years prior to diagnosis of early gastric cancer

JGH Open ◽  
2020 ◽  
Author(s):  
Yo Fujimoto ◽  
Yasumi Katayama ◽  
Yoshinori Gyotoku ◽  
Ryosuke Oura ◽  
Ikuhiro Kobori ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Early Gastric Cancer ◽  
Risk Score ◽  
Predictive Value

scholarly journals Breath analysis based early gastric cancer classification from deep stacked sparse autoencoder neural network

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Muhammad Aqeel Aslam ◽  
Cuili Xue ◽  
Yunsheng Chen ◽  
Amin Zhang ◽  
Manhua Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Gastric Cancer ◽  
Feature Extraction ◽  
Early Gastric Cancer ◽  
Breath Analysis ◽  
Cancer Classification ◽  
Unlabeled Data ◽  
Softmax Classifier ◽  
Sparse Autoencoder ◽  
Stacked Sparse Autoencoder

AbstractDeep learning is an emerging tool, which is regularly used for disease diagnosis in the medical field. A new research direction has been developed for the detection of early-stage gastric cancer. The computer-aided diagnosis (CAD) systems reduce the mortality rate due to their effectiveness. In this study, we proposed a new method for feature extraction using a stacked sparse autoencoder to extract the discriminative features from the unlabeled data of breath samples. A Softmax classifier was then integrated to the proposed method of feature extraction, to classify gastric cancer from the breath samples. Precisely, we identified fifty peaks in each spectrum to distinguish the EGC, AGC, and healthy persons. This CAD system reduces the distance between the input and output by learning the features and preserve the structure of the input data set of breath samples. The features were extracted from the unlabeled data of the breath samples. After the completion of unsupervised training, autoencoders with Softmax classifier were cascaded to develop a deep stacked sparse autoencoder neural network. In last, fine-tuning of the developed neural network was carried out with labeled training data to make the model more reliable and repeatable. The proposed deep stacked sparse autoencoder neural network architecture exhibits excellent results, with an overall accuracy of 98.7% for advanced gastric cancer classification and 97.3% for early gastric cancer detection using breath analysis. Moreover, the developed model produces an excellent result for recall, precision, and f score value, making it suitable for clinical application.

scholarly journals Fusing Logical Relationship Information of Text in Neural Network for Text Classification

2020 ◽  
Vol 2020 ◽  
pp. 1-16 ◽  
Author(s):  
Heyong Wang ◽  
Dehang Zeng
Keyword(s):  
Neural Network ◽  
Text Classification ◽  
Information Science ◽  
Classification Algorithms ◽  
Human Beings ◽  
Central Idea ◽  
Logical Relationship ◽  
The Relationship ◽  
Different Parts ◽  
Better Than

With the development of computer science and information science, text classification technology has been greatly developed and its application scenarios have been widened. In traditional process of text classification, the existing method will lose much logical relationship information of text. The logical relationship information of a text refers to the relationship information among different logical parts of the text, such as title, abstract, and body. When human beings are reading, they will take title as an important part to remind the central idea of the article, abstract as a brief summary of the content of the article, and body as a detailed description of the article. In most of the text classification studies, researchers concern more about the relationship among words (word frequency, semantics, etc.) and neglect the logical relationship information of text. It will lose information about the relationship among different parts (title, body, etc.) and have an influence on the performance of text classification. Therefore, we propose a text classification algorithm—fusing the logical relationship information of text in neural network (FLRIOTINN), which complements the logical relationship information into text classification algorithms. Experiments show that the effect of FLRIOTINN is better than the conventional backpropagation neural networks which does not consider the logical relationship information of text.

Laparoscopy-Assisted Pylorus-Preserving Gastrectomy Is Better Than Laparoscopy-Assisted Distal Gastrectomy for Middle-Third Early Gastric Cancer

2014 ◽  
Vol 259 (3) ◽  
pp. 485-493 ◽  
Author(s):  
Yun-Suhk Suh ◽  
Dong-Seok Han ◽  
Seong-Ho Kong ◽  
Sebastianus Kwon ◽  
Cheong-Il Shin ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Early Gastric Cancer ◽  
Distal Gastrectomy ◽  
Better Than

scholarly journals Convolutional neural network for the diagnosis of early gastric cancer based on magnifying narrow band imaging

2019 ◽  
Vol 23 (1) ◽  
pp. 126-132 ◽  
Author(s):  
Lan Li ◽  
Yishu Chen ◽  
Zhe Shen ◽  
Xuequn Zhang ◽  
Jianzhong Sang ◽  
...  
Keyword(s):  
Neural Network ◽  
Gastric Cancer ◽  
Early Gastric Cancer ◽  
Convolutional Neural Network ◽  
Narrow Band ◽  
Narrow Band Imaging

scholarly journals The Use of CD44 Variant 9 and Ki-67 Combination Can Predicts Prognosis Better Than Their Single Use in Early Gastric Cancer

2019 ◽  
Vol 51 (4) ◽  
pp. 1411-1419 ◽  
Author(s):  
Se-Il Go ◽  
Gyung Hyuck Ko ◽  
Won Sup Lee ◽  
Jeong-Hee Lee ◽  
Sang-Ho Jeong ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Early Gastric Cancer ◽  
Ki 67 ◽  
Cd44 Variant ◽  
Single Use ◽  
Better Than

scholarly journals Comparison of postoperative lymphocytes and interleukins between laparoscopy-assisted and open radical gastrectomy for early gastric cancer

2018 ◽  
Vol 47 (1) ◽  
pp. 303-310 ◽  
Author(s):  
Xiang Xia ◽  
Zizhen Zhang ◽  
Jia Xu ◽  
Gang Zhao ◽  
Fengrong Yu
Keyword(s):  
Gastric Cancer ◽  
T Cells ◽  
Early Gastric Cancer ◽  
Cd8 T Cells ◽  
Inflammatory Factors ◽  
Radical Gastrectomy ◽  
The Third ◽  
Estimated Blood Loss ◽  
Laparoscopic Assisted ◽  
Lymphocytes Subsets

Objective This study aimed to study the effects of laparoscopic-assisted radical gastrectomy (LAG) and open radical gastrectomy (OG) on immune function and inflammatory factors in patients with early gastric cancer. Methods Seventy-five patients with pT1N0M0 gastric cancer in Ren Ji Hospital from August 2017 to January 2018 were studied. Lymphocytes subsets and interleukins were compared preoperatively and on the third postoperative day (POD3) and seventh postoperative day (POD7). Results There were no significant differences in age, sex, body mass index, duration of the operation, estimated blood loss, total gastrectomy rate, postoperative first fluid diet, and the levels of preoperative lymphocytes subsets and interleukins between the two groups. The number of CD4+ T cells and the CD4+/CD8+ ratio in the LAG group were significantly higher than those in the OG group on POD3. However, the number of CD8+ T cells, and interleukin-6 and interleukin-8 levels in the LAG group were significantly lower than those in the OG group on POD3. Conclusions Laparoscopy can effectively reduce the levels of inflammatory factors and has less effect on the immune system than OG.

Sign in / Sign up

Export Citation Format

Share Document