Automatic detection of small bowel lesions with different bleeding risks based on deep learning models.

BACKGROUND: Deep learning provides an efficient automatic image recognition method for small bowel (SB) capsule endoscopy (CE) that can assist physicians in diagnosis. However, the existing deep learning models present some unresolved challenges. AIM: To propose a novel and effective classification and detection model to automatically identify various SB lesions and their bleeding risks, and label the lesions accurately so as to enhance the diagnostic efficiency of physicians and the ability to identify high-risk bleeding groups. METHODS: The proposed model represents a two-stage method that combined image classification with object detection. First, we utilized the improved ResNet-50 classification model to classify endoscopic images into SB lesion images, normal SB mucosa images, and invalid images. Then, the improved YOLO-V5 detection model was utilized to detect the type of lesion and its risk of bleeding, and the location of the lesion was marked. We constructed training and testing sets and compared model-assisted reading with physician reading. RESULTS: The accuracy of the model constructed in this study reached 98.96%, which was higher than the accuracy of other systems using only a single module. The sensitivity, specificity, and accuracy of the model-assisted reading detection of all images were 99.17%, 99.92%, and 99.86%, which were significantly higher than those of the endoscopists' diagnoses. The image processing time of the model was 48 ms/image, and the image processing time of the physicians was 0.40 ± 0.24 s/image (P < 0.001). CONCLUSION: The deep learning model of image classification combined with object detection exhibits a satisfactory diagnostic effect on a variety of SB lesions and their bleeding risks in CE images, which enhances the diagnostic efficiency of physicians and improves the ability of physicians to identify high-risk bleeding groups.

[Comparison of Rhesus boxes in Hans and Uighurs].

OBJECTIVE: To study the difference and similarity between Hans and Uighurs in regard to Rhesus box and its significance. METHODS: The sequence specific primers of upstream, downstream and hybrid Rhesus boxes were designed on the basis of RHD gene sequence. The upstream, downstream and hybrid Rhesus boxes were determined by polymerase chain reaction-sequence specific primer(PCP-SSP) and mismatched PCR. RESULTS: The percentage of RHD-/RHD-, RHD+/RHD- and RHD+/RHD+ genotypes ascertained in the unrelated Hans with RhD(-) were 61.40%, 34.21% and 4.39% respectively, while those in the unrelated Chinese Uighurs with RhD(-) were 94.44%, 2.78% and 2.78% respectively. Furthermore, all 6 cases of some other minorities were RHD-/RHD- types. The percentage of RHD-/RHD- and RHD+/RHD- genotypes ascertained in the unrelated Chinese Uighurs were significantly higher than those in Chinese Hans (P < 0.01), whereas no statistically significant difference in the percentage of RHD+/RDH+ genotype between the two groups was observed (P > 0.05). CONCLUSION: The Rh blood group of Uighurs in Xingjiang possesses both Oriental and Caucasian characteristics, which embodies a special ethnical aspect of the Chinese nation and is in accord with the anthropologic research results.