Automated detection of small bowel lesions based on capsule endoscopy using deep learning algorithm.

BACKGROUND: In order to overcome the challenges of lesion detection in capsule endoscopy (CE), we improved the YOLOv5-based deep learning algorithm and established the CE-YOLOv5 algorithm to identify small bowel lesions captured by CE. METHODS: A total of 124,678 typical abnormal images from 1,452 patients were enrolled to train the CE-YOLOv5 model. Then 298 patients with suspected small bowel lesions detected by CE were prospectively enrolled in the testing phase of the study. Small bowel images and videos from the above 298 patients were interpreted by the experts, non-experts and CE-YOLOv5, respectively. RESULTS: The sensitivity of CE-YOLOv5 in diagnosing vascular lesions, ulcerated/erosive lesions, protruding lesions, parasite, diverticulum, active bleeding and villous lesions based on CE videos was 91.9 %, 92.2 %, 91.4 %, 93.1 %, 93.3 %, 95.1 %, and 100 % respectively. Furthermore, CE-YOLOv5 achieved specificity and accuracy of more than 90 % for all lesions. Compared with experts, the CE-YOLOv5 showed comparable overall sensitivity, specificity and accuracy (all P > 0.05). Compared with non-experts, the CE-YOLOv5 showed significantly higher overall sensitivity (P < 0.0001) and overall accuracy (P < 0.0001), and a moderately higher overall specificity (P = 0.0351). Furthermore, the time for AI-reading (5.62 ± 2.81 min) was significantly shorter than that for the other two groups (both P < 0.0001). CONCLUSIONS: CE-YOLOv5 diagnosed small bowel lesions in CE videos with high sensitivity, specificity and accuracy, providing a reliable approach for automated lesion detection in real-world clinical practice.

LncRNA-AK007111 affects airway inflammation in asthma via the regulation of mast cell function.

Long noncoding RNAs (lncRNAs) are involved in gene transcription and pathophysiological processes of human diseases. Multiple lncRNAs have been shown to play important roles in the occurrence and development of asthma. This study aimed to explore the role of a novel lncRNA, lncRNA-AK007111, in asthma. Overexpression of lncRNA-AK007111 was induced in a mouse model of asthma via viral transfection, followed by the collection of alveolar lavage fluid and lung tissue for the detection of relevant inflammatory factors and pathological analysis of lung sections. Pulmonary resistance and respiratory dynamic compliance were measured using an animal pulmonary function analyzer. The number of mast cells sensitized by immunofluorescence was detected at the cellular level. The degree of degranulation of lncRNA-AK007111 after its knockdown was determined by detecting the level of ß-hexosaminidase that was released and quantifying IL-6 and TNF-α using ELISA in a model of RBL-2H3 cells activated by immunoglobulin E plus antigen. Finally, we observed the migration ability of mast cells under a microscope. The results showed that in ovalbumin-sensitized mice, the upregulation of lncRNA-AK007111 promoted the infiltration of inflammatory cells in lung tissue, increased the number of total cells, eosinophils, and mast cells, upregulated IL-5 and IL-6 levels, and increased airway hyper-reactivity. Downregulation of lncRNA-AK007111 decreased the degranulation ability of IgE/Ag-activated mast cells and inhibited the expression of IL-6 and TNF-α; moreover, the migration ability of mast cells was significantly weakened. In conclusion, our study revealed that lncRNA-AK007111 plays an important role in asthma by modulating mast cell-related functions.

[Consideration on Animal Experiment in PET/CT].

PET/CT imaging can reflect the physiological metabolic process in living body which is the model experiment incapable to simulate. Animal experiment may be considered for systematic validation of PET/CT products. The obtained research data can be used to evaluate the feasibility, effectiveness and safety of PET/CT products, and be submitted as supporting documents for research data or clinical evaluation data when doing product registration or alteration registration. In this study, the functions and advantages of animal experiments were expounded, and relevant research cases were given as well as the issues that should be paid attention to. It can be a reference for the validation and review of PET/CT products.

Phase Contrast Image-Based Rapid Antimicrobial Susceptibility Testing of Bacteria in Liquid Culture Media.

Currently, the world is facing the problem of bacterial resistance, which threatens public health, and bacterial antimicrobial susceptibility testing (AST) plays an important role in biomedicine, dietary safety and aquaculture. Traditional AST methods take a long time, usually 16-24 h, and cannot meet the demand for rapid diagnosis in the clinic, so rapid AST methods are needed to shorten the detection time. In this study, by using an in-house built centrifuge to centrifuge bacteria in a liquid medium onto the inner wall of the bottom surface of a counting plate, and using a phase contrast microscope to track bacterial growth under the effect of different antibiotic concentrations, the results of the minimum inhibitory concentration (MIC) of bacteria under the effect of antibiotics can be obtained in as early as 4 h. We used a combination of E. coli and tigecycline and obtained MIC results that were consistent with those obtained using the gold standard broth micro-dilution method, demonstrating the validity of our method; due to the time advantage, the complete set can be used in the future for point of care and clinical applications, helping physicians to quickly obtain the MIC used to inhibit bacterial growth.

Recovery of lanthanum cations by functionalized magnetic multi-walled carbon nanotube bundles.

Rare-earth elements (REE), including La, are critical raw materials in many technological advancements. Collection of physically adsorbed REEs on clay minerals can be realized first by ion-exchange leaching, followed by adsorption enrichment. Ever increasing demand and limited resources of REEs have fueled the development of nanostructured adsorbents. In this paper, multi-walled carbon nanotubes (MWCNTs) were purified using concentrated H2SO4 and HNO3, then coupled with magnetic Fe3O4 nanoparticles to make low concentration La ion extraction from water possible. The MWCNT@Fe3O4 composites were further crosslinked with 0.1 wt% epichlorohydrin and functionalized with 0.5 wt% carbon disulfide to achieve a La3+ adsorption capacity of 23.23 mg g-1. We fully probed the morphology, crystallinity, chemical composition, and magnetic properties of the as-prepared adsorbent by scanning/transmission electron microscopy, X-ray diffractometry, X-ray photoelectron spectroscopy, vibrating-sample magnetometry, and thermal gravimetry. These results indicated that the MWCNT@Fe3O4 nanohybrid may be a promising candidate for recovering La ions from aqueous solutions.