A newly developed deep learning-based system for automatic detection and classification of small bowel lesions during double-balloon enteroscopy examination.

BACKGROUND: Double-balloon enteroscopy (DBE) is a standard method for diagnosing and treating small bowel disease. However, DBE may yield false-negative results due to oversight or inexperience. We aim to develop a computer-aided diagnostic (CAD) system for the automatic detection and classification of small bowel abnormalities in DBE. DESIGN AND METHODS: A total of 5201 images were collected from Renmin Hospital of Wuhan University to construct a detection model for localizing lesions during DBE, and 3021 images were collected to construct a classification model for classifying lesions into four classes, protruding lesion, diverticulum, erosion & ulcer and angioectasia. The performance of the two models was evaluated using 1318 normal images and 915 abnormal images and 65 videos from independent patients and then compared with that of 8 endoscopists. The standard answer was the expert consensus. RESULTS: For the image test set, the detection model achieved a sensitivity of 92% (843/915) and an area under the curve (AUC) of 0.947, and the classification model achieved an accuracy of 86%. For the video test set, the accuracy of the system was significantly better than that of the endoscopists (85% vs. 77 ± 6%, p < 0.01). For the video test set, the proposed system was superior to novices and comparable to experts. CONCLUSIONS: We established a real-time CAD system for detecting and classifying small bowel lesions in DBE with favourable performance. ENDOANGEL-DBE has the potential to help endoscopists, especially novices, in clinical practice and may reduce the miss rate of small bowel lesions.

IKAROS and CK2 regulate expression of BCL-XL and chemosensitivity in high-risk B-cell acute lymphoblastic leukemia.

High-risk B-cell acute lymphoblastic leukemia (B-ALL) is an aggressive disease, often characterized by resistance to chemotherapy. A frequent feature of high-risk B-ALL is loss of function of the IKAROS (encoded by the IKZF1 gene) tumor suppressor. Here, we report that IKAROS regulates expression of the BCL2L1 gene (encodes the BCL-XL protein) in human B-ALL. Gain-of-function and loss-of-function experiments demonstrate that IKAROS binds to the BCL2L1 promoter, recruits histone deacetylase HDAC1, and represses BCL2L1 expression via chromatin remodeling. In leukemia, IKAROS' function is impaired by oncogenic casein kinase II (CK2), which is overexpressed in B-ALL. Phosphorylation by CK2 reduces IKAROS binding and recruitment of HDAC1 to the BCL2L1 promoter. This results in a loss of IKAROS-mediated repression of BCL2L1 and increased expression of BCL-XL. Increased expression of BCL-XL and/or CK2, as well as reduced IKAROS expression, are associated with resistance to doxorubicin treatment. Molecular and pharmacological inhibition of CK2 with a specific inhibitor CX-4945, increases binding of IKAROS to the BCL2L1 promoter and enhances IKAROS-mediated repression of BCL2L1 in B-ALL. Treatment with CX-4945 increases sensitivity to doxorubicin in B-ALL, and reverses resistance to doxorubicin in multidrug-resistant B-ALL. Combination treatment with CX-4945 and doxorubicin show synergistic therapeutic effects in vitro and in preclinical models of high-risk B-ALL. Results reveal a novel signaling network that regulates chemoresistance in leukemia. These data lay the groundwork for clinical testing of a rationally designed, targeted therapy that combines the CK2 inhibitor, CX-4945, with doxorubicin for the treatment of hematopoietic malignancies.

Factors influencing risk perception and nosocomial infection prevention practices of frontline nurses during the COVID-19 pandemic.

BACKGROUND: During the coronavirus disease 2019 (COVID-19) pandemic, exploring factors influencing nosocomial infection among frontline nurses may provide evidence to optimize prevention strategies in hospitals. METHOD: A large-scale online questionnaire survey of nurses' state-trait anxiety, job burnout, risk perception, workplace safety perception, knowledge about nosocomial infection, and preventive practices was conducted with 2795 frontline nurses working in the COVID-19 wards of six hospitals in Hubei Province, China, from February 1 to April 1, 2020. The questionnaire data were analyzed using the structural equation modeling (SEM) method to reveal the mechanisms influencing nurses' risk perception and preventive practices related to nosocomial COVID-19 infection. RESULTS: A model of the factors that influence nurses' risk perception and preventive practices regarding nosocomial COVID-19 infection was established. The model verified hypotheses regarding the impact of nurses' risk perception and preventive practices. Notably, the hypothesis that risk perception has an impact on nurses' preventive practices regarding nosocomial infection is not valid. Moreover, different marital and educational conditions are associated with significant differences in the impact of state anxiety on the execution of preventive practices, the impact of workplace safety perceptions on risk perception, and the impact of workplace safety perceptions on the execution of preventive practices. The effect of state anxiety on preventive practices differed significantly with different durations of work experience. CONCLUSIONS: According to the results of the influencing factor model, promoting the quality of training on nosocomial infection, meliorating workplace safety, and conducting timely and effective psychological interventions would aid in improving nurses' preventive practices. Meliorating workplace safety and easing state anxiety would be beneficial to reduce nurses' risk perception. These strategies are conducive to the optimization of policies for preventing nosocomial COVID-19 infections and similar infectious diseases.

[Cyclooxygenase-2 -765G>C polymorphism and susceptibility to colorectal cancer: a meta-analysis].

OBJECTIVE: To explore the correlation between polymorphism of cyclooxygenase-2 (COX-2) -765G>C and susceptibility to colorectal cancer. METHODS: All eligible case-control studies published up to March 2013 were searched out from PubMed, EMABSE, CJFD, CBM, CNKI, VIP and WanFang databases, while 99 articles were concluded. Two reviewers independently identified the literature according to inclusion and exclusion criteria. Meta-analysis was performed using RevMan 5.1 and Stata 12.0 software. RESULTS: A total of eleven studies comprising 3432 cases and 5286 controls were finally included. The included studies showed good homogeneity in the three genetic models, except the model of GC/GG genotype (I(2) = 52%, P = 0.03). Overall, there were no significant association between polymorphism of COX-2-765G>C and the susceptibility to colorectal cancer (dominant model: (GC+CC)/GG: OR = 1.08, 95%CI:0.96-1.21; recessive model:CC/(GC+GG): OR = 1.09, 95%CI:0.76-1.56; GC/GG: OR = 1.05, 95%CI:0.87-1.28; CC/GG: OR = 1.11, 95%CI:0.77-1.60). In stratification analysis by ethnicity, we observed that the polymorphism of COX-2 -765G>C could increase the susceptibility to colorectal cancer among yellow populations ((GC+CC)/GG: OR = 1.41, 95%CI:1.15-1.75; GC/ GG: OR = 1.48, 95%CI:1.15-1.90), but there was no significant association found among Caucasian populations. CONCLUSION: This meta-analysis suggested that the polymorphism of COX-2 -765G>C may increase the susceptibility to colorectal cancer in the yellow population.

Early Triage of Critically Ill Adult Patients With Mushroom Poisoning: Machine Learning Approach.

BACKGROUND: Early triage of patients with mushroom poisoning is essential for administering precise treatment and reducing mortality. To our knowledge, there has been no established method to triage patients with mushroom poisoning based on clinical data. OBJECTIVE: The purpose of this work was to construct a triage system to identify patients with mushroom poisoning based on clinical indicators using several machine learning approaches and to assess the prediction accuracy of these strategies. METHODS: In all, 567 patients were collected from 5 primary care hospitals and facilities in Enshi, Hubei Province, China, and divided into 2 groups; 322 patients from 2 hospitals were used as the training cohort, and 245 patients from 3 hospitals were used as the test cohort. Four machine learning algorithms were used to construct the triage model for patients with mushroom poisoning. Performance was assessed using the area under the receiver operating characteristic curve (AUC), decision curve, sensitivity, specificity, and other representative statistics. Feature contributions were evaluated using Shapley additive explanations. RESULTS: Among several machine learning algorithms, extreme gradient boosting (XGBoost) showed the best discriminative ability in 5-fold cross-validation (AUC=0.83, 95% CI 0.77-0.90) and the test set (AUC=0.90, 95% CI 0.83-0.96). In the test set, the XGBoost model had a sensitivity of 0.93 (95% CI 0.81-0.99) and a specificity of 0.79 (95% CI 0.73-0.85), whereas the physicians' assessment had a sensitivity of 0.86 (95% CI 0.72-0.95) and a specificity of 0.66 (95% CI 0.59-0.73). CONCLUSIONS: The 14-factor XGBoost model for the early triage of mushroom poisoning can rapidly and accurately identify critically ill patients and will possibly serve as an important basis for the selection of treatment options and referral of patients, potentially reducing patient mortality and improving clinical outcomes.

Exploring the Interdisciplinary Nature of Precision Medicine：Network Analysis and Visualization.

BACKGROUND: Interdisciplinary research is an important feature of precision medicine. However, the accurate cross-disciplinary status of precision medicine is still unclear. OBJECTIVE: The aim of this study is to present the nature of interdisciplinary collaboration in precision medicine based on co-occurrences and social network analysis. METHODS: A total of 7544 studies about precision medicine, published between 2010 and 2019, were collected from the Web of Science database. We analyzed interdisciplinarity with descriptive statistics, co-occurrence analysis, and social network analysis. An evolutionary graph and strategic diagram were created to clarify the development of streams and trends in disciplinary communities. RESULTS: The results indicate that 105 disciplines are involved in precision medicine research and cover a wide range. However, the disciplinary distribution is unbalanced. Current cross-disciplinary collaboration in precision medicine mainly focuses on clinical application and technology-associated disciplines. The characteristics of the disciplinary collaboration network are as follows: (1) disciplinary cooperation in precision medicine is not mature or centralized; (2) the leading disciplines are absent; (3) the pattern of disciplinary cooperation is mostly indirect rather than direct. There are 7 interdisciplinary communities in the precision medicine collaboration network; however, their positions in the network differ. Community 4, with disciplines such as genetics and heredity in the core position, is the most central and cooperative discipline in the interdisciplinary network. This indicates that Community 4 represents a relatively mature direction in interdisciplinary cooperation in precision medicine. Finally, according to the evolution graph, we clearly present the development streams of disciplinary collaborations in precision medicine. We describe the scale and the time frame for development trends and distributions in detail. Importantly, we use evolution graphs to accurately estimate the developmental trend of precision medicine, such as biological big data processing, molecular imaging, and widespread clinical applications. CONCLUSIONS: This study can help researchers, clinicians, and policymakers comprehensively understand the overall network of interdisciplinary cooperation in precision medicine. More importantly, we quantitatively and precisely present the history of interdisciplinary cooperation and accurately predict the developing trends of interdisciplinary cooperation in precision medicine.

Intellectual Structure and Evolutionary Trends of Precision Medicine Research: Coword Analysis.

BACKGROUND: Precision medicine (PM) is playing a more and more important role in clinical practice. In recent years, the scale of PM research has been growing rapidly. Many reviews have been published to facilitate a better understanding of the status of PM research. However, there is still a lack of research on the intellectual structure in terms of topics. OBJECTIVE: This study aimed to identify the intellectual structure and evolutionary trends of PM research through the application of various social network analysis and visualization methods. METHODS: The bibliographies of papers published between 2009 and 2018 were extracted from the Web of Science database. Based on the statistics of keywords in the papers, a coword network was generated and used to calculate network indicators of both the entire network and local networks. Communities were then detected to identify subdirections of PM research. Topological maps of networks, including networks between communities and within each community, were drawn to reveal the correlation structure. An evolutionary graph and a strategic graph were finally produced to reveal research venation and trends in discipline communities. RESULTS: The results showed that PM research involves extensive themes and, overall, is not balanced. A minority of themes with a high frequency and network indicators, such as Biomarkers, Genomics, Cancer, Therapy, Genetics, Drug, Target Therapy, Pharmacogenomics, Pharmacogenetics, and Molecular, can be considered the core areas of PM research. However, there were five balanced theme directions with distinguished status and tendencies: Cancer, Biomarkers, Genomics, Drug, and Therapy. These were shown to be the main branches that were both focused and well developed. Therapy, though, was shown to be isolated and undeveloped. CONCLUSIONS: The hotspots, structures, evolutions, and development trends of PM research in the past ten years were revealed using social network analysis and visualization. In general, PM research is unbalanced, but its subdirections are balanced. The clear evolutionary and developmental trend indicates that PM research has matured in recent years. The implications of this study involving PM research will provide reasonable and effective support for researchers, funders, policymakers, and clinicians.