Identification of visible lesions during surveillance endoscopy for Barrett's esophagus: a video-based survey study

Robustness evaluation of deep neural networks for endoscopic image analysis: Insights and strategies

Computer-aided detection and diagnosis systems (CADe/CADx) in endoscopy are commonly trained using high-quality imagery, which is not representative for the heterogeneous input typically encountered in clinical practice. In endoscopy, the image quality heavily relies on both the skills and experience of the endoscopist and the specifications of the system used for screening. Factors such as poor illumination, motion blur, and specific post-processing settings can significantly alter the quality and general appearance of these images. This so-called domain gap between the data used for developing the system and the data it encounters after deployment, and the impact it has on the performance of deep neural networks (DNNs) supportive endoscopic CAD systems remains largely unexplored. As many of such systems, for e.g. polyp detection, are already being rolled out in clinical practice, this poses severe patient risks in particularly community hospitals, where both the imaging equipment and experience are subject to considerable variation. Therefore, this study aims to evaluate the impact of this domain gap on the clinical performance of CADe/CADx for various endoscopic applications. For this, we leverage two publicly available data sets (KVASIR-SEG and GIANA) and two in-house data sets. We investigate the performance of commonly-used DNN architectures under synthetic, clinically calibrated image degradations and on a prospectively collected dataset including 342 endoscopic images of lower subjective quality. Additionally, we assess the influence of DNN architecture and complexity, data augmentation, and pretraining techniques for improved robustness. The results reveal a considerable decline in performance of 11.6% (±1.5) as compared to the reference, within the clinically calibrated boundaries of image degradations. Nevertheless, employing more advanced DNN architectures and self-supervised in-domain pre-training effectively mitigate this drop to 7.7% (±2.03). Additionally, these enhancements yield the highest performance on the manually collected test set including images with lower subjective quality. By comprehensively assessing the robustness of popular DNN architectures and training strategies across multiple datasets, this study provides valuable insights into their performance and limitations for endoscopic applications. The findings highlight the importance of including robustness evaluation when developing DNNs for endoscopy applications and propose strategies to mitigate performance loss.

Construction of the Interaction Network of Hub Genes in the Progression of Barrett's Esophagus to Esophageal Adenocarcinoma

Introduction

Esophageal adenocarcinoma (EAC) is one of the histologic types of esophageal cancer with a poor prognosis. The majority of EAC originate from Barrett's esophagus (BE). There are few studies focusing on the dynamic progression of BE to EAC.

Methods

R software was used to analyze differentially expressed genes (DEGs) based on RNA-seq data of 94 normal esophageal squamous epithelial (NE) tissues, 113 BE tissues and 147 EAC tissues. The overlapping genes of DEGs between BE and EAC were analyzed by Venn diagram tool. The hub genes were selected by Cytoscape software based on the protein-protein interaction network of the overlapping genes using STRING database. The functional analysis of hub genes was performed by R software and the protein expression was identified by immunohistochemistry.

Results

In the present study, we found a large degree of genetic similarity between BE and EAC, and further identified seven hub genes (including COL1A1, TGFBI, MMP1, COL4A1, NID2, MMP12, CXCL1) which were all progressively upregulated in the progression of NE-BE-EAC. We have preliminarily uncovered the probable molecular mechanisms of these hub genes in disease development and constructed the ceRNA regulatory network of hub genes. More importantly, we explored the possibility of hub genes as biomarkers in the disease progression of NE-BE-EAC. For example, TGFBI can be used as biomarkers to predict the prognosis of EAC patients. COL1A1, NID2 and COL4A1 can be used as biomarkers to predict the response to immune checkpoint blockade (ICB) therapy. We also constructed a disease progression risk model for NE-BE-EAC based on CXCL1, MMP1 and TGFBI. Finally, the results of drug sensitivity analysis based on hub genes showed that drugs such as PI3K inhibitor TGX221, bleomycin, PKC inhibitor Midostaurin, Bcr-Abl inhibitor Dasatinib, HSP90 inhibitor 17-AAG, and Docetaxel may be potential candidates to inhibit the progression of BE to EAC.

Conclusion

This study is based on a large number of clinical samples with high credibility, which is useful for revealing the probable carcinogenic mechanism of BE to EAC and developing new clinical treatment strategies.