Enhanced segmentation of gastrointestinal polyps from capsule endoscopy images with artifacts using ensemble learning.

BACKGROUND: Endoscopy artifacts are widespread in real capsule endoscopy (CE) images but not in high-quality standard datasets. AIM: To improve the segmentation performance of polyps from CE images with artifacts based on ensemble learning. METHODS: We collected 277 polyp images with CE artifacts from 5760 h of videos from 480 patients at Guangzhou First People's Hospital from January 2016 to December 2019. Two public high-quality standard external datasets were retrieved and used for the comparison experiments. For each dataset, we randomly segmented the data into training, validation, and testing sets for model training, selection, and testing. We compared the performance of the base models and the ensemble model in segmenting polyps from images with artifacts. RESULTS: The performance of the semantic segmentation model was affected by artifacts in the sample images, which also affected the results of polyp detection by CE using a single model. The evaluation based on real datasets with artifacts and standard datasets showed that the ensemble model of all state-of-the-art models performed better than the best corresponding base learner on the real dataset with artifacts. Compared with the corresponding optimal base learners, the intersection over union (IoU) and dice of the ensemble learning model increased to different degrees, ranging from 0.08% to 7.01% and 0.61% to 4.93%, respectively. Moreover, in the standard datasets without artifacts, most of the ensemble models were slightly better than the base learner, as demonstrated by the IoU and dice increases ranging from -0.28% to 1.20% and -0.61% to 0.76%, respectively. CONCLUSION: Ensemble learning can improve the segmentation accuracy of polyps from CE images with artifacts. Our results demonstrated an improvement in the detection rate of polyps with interference from artifacts.

Characterizing the composition of intestinal microflora by 16S rRNA gene sequencing.

BACKGROUND: This study determined the composition and diversity of intestinal microflora in patients with colorectal adenoma (CRA), which may provide precedence for investigating the role of intestinal microflora in the pathogenesis of colorectal tumors, the composition of intestinal microflora closely related to CRA, and further validating the possibility of intestinal flora as a biomarker of CRA. AIM: To study the relationship between intestinal microflora and CRA. METHODS: This is a prospective control case study from October 2014 to June 2015 involving healthy volunteers and patients with advanced CRA. High-throughput sequencing and bioinformatics analysis were used to investigate the composition and diversity of intestinal microflora in 36 healthy subjects and 49 patients with advanced CRA. Endpoints measured were operational taxonomic units of intestinal flora, as well as their abundance and diversity (α and ß types). RESULTS: In this study, the age, gender, body mass index, as well as location between controls and patients had no significant differences. The mucosa-associated gut microbiota diversity and bacterial distribution in healthy controls and colorectal adenomas were similar. The operational taxonomic unit, abundance, and α and ß diversity were all reduced in patients with CRA compared to controls. At the phylum level, the composition of intestinal microflora was comparable between patients and controls, but the abundance of Proteobacteria was increased, and Firmicutes and Bacteroides were significantly decreased (P < 0.05). The increase in Halomonadaceae and Shewanella algae, and reduction in Coprococcus and Bacteroides ovatus, could serve as biomarkers of CRA. High-throughput sequencing confirms the special characteristics and diversity of intestinal microflora in healthy controls and patients with CRA. CONCLUSION: The diversity of intestinal microflora was decreased in patients with CRA. An increase in Halomonadaceae and Shewanella algae are markers of CRA.