Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images (vol 14, 7847, 2024)

被引：1

作者：

Zhang, Kezhi ^{[1
]}

Wang, Haibao ^{[1
]}

Cheng, Yaru ^{[2
]}

Liu, Hongyan ^{[2
]}

Gong, Qi ^{[2
]}

Zeng, Qian ^{[1
]}

Zhang, Tao ^{[1
]}

Wei, Guoqiang ^{[1
,3
]}

Wei, Zhi ^{[2
]}

Chen, Dong ^{[1
]}

机构：

[1] Nanning Normal Univ, Sch Phys & Elect, Guangxi Key Lab Informat Funct Mat & Intelligent I, 175 Mingxiu East Rd, Nanning 530001, Guangxi, Peoples R China

[2] Shandong Second Prov Gen Hosp, Dept Gastroenterol, 4 Duan Xing West Rd, Jinan 250022, Shandong, Peoples R China

[3] Hunan Coll Informat, Sch Elect Engn, Changsha 410200, Hunan, Peoples R China

来源：

SCIENTIFIC REPORTS | 2024年 / 14卷 / 01期

基金：

中国国家自然科学基金;

关键词：

Deep learning; Early gastric cancer; Gastroscopic images; Mask R-CNN;

D O I：

10.1038/s41598-024-59794-x

中图分类号：

O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

Gastric cancer is a highly prevalent disease that poses a serious threat to public health. In clinical practice, gastroscopy is frequently used by medical practitioners to screen for gastric cancer. However, the symptoms of gastric cancer at different stages of advancement vary significantly, particularly in the case of early gastric cancer (EGC). The manifestations of EGC are often indistinct, leading to a detection rate of less than 10%. In recent years, researchers have focused on leveraging deep learning algorithms to assist medical professionals in detecting EGC and thereby improve detection rates. To enhance the ability of deep learning to detect EGC and segment lesions in gastroscopic images, an Improved Mask R-CNN (IMR-CNN) model was proposed. This model incorporates a "Bi-directional feature extraction and fusion module" and a "Purification module for feature channel and space" based on the Mask R-CNN (MR-CNN). Our study includes a dataset of 1120 images of EGC for training and validation of the models. The experimental results indicate that the IMR-CNN model outperforms the original MR-CNN model, with Precision, Recall, Accuracy, Specificity and F1-Score values of 92.9%, 95.3%, 93.9%, 92.5% and 94.1%, respectively. Therefore, our proposed IMR-CNN model has superior detection and lesion segmentation capabilities and can effectively aid doctors in diagnosing EGC from gastroscopic images.

引用

页数：1