COVID19 Diagnosis Using Chest X-rays and Transfer Learning

A pandemic of respiratory illnesses from a novel coronavirus known as Sars-CoV-2 has swept across the globe since December of 2019. This is calling upon the research community including medical imaging to provide effective tools for use in combating this virus. Research in biomedical imaging of viral patients is already very active with machine learning models being created for diagnosing Sars-CoV-2 infections in patients using CT scans and chest x-rays. We aim to build upon this research. Here we used a transfer-learning approach to develop models capable of diagnosing COVID19 from chest x-ray. For this work we compiled a dataset of 112120 negative images from the Chest X-Ray 14 and 2725 positive images from public repositories. We tested multiple models, including logistic regression and random forest and XGBoost with and without principal components analysis, using five-fold cross-validation to evaluate recall, precision, and f1-score. These models were compared to a pre-trained deep-learning model for evaluating chest x-rays called COVID-Net. Our best model was XGBoost with principal components with a recall, precision, and f1-score of 0.692, 0.960, 0.804 respectively. This model greatly outperformed COVID-Net which scored 0.987, 0.025, 0.048. This model, with its high precision and reasonable sensitivity, would be most useful as "rule-in" test for COVID19. Though it outperforms some chemical assays in sensitivity, this model should be studied in patients who would not ordinarily receive a chest x-ray before being used for screening. CCS CONCEPTSO_LILife and Medical Sciences * Machine Learning * Artificial Intelligence C_LI Reference formatJonathan Stubblefield, Jason Causey, Dakota Dale, Jake Qualls, Emily Bellis, Jennifer Fowler, Karl Walker and Xiuzhen Huang. 2022. COVID19 Diagnosis Using Chest X-Rays and Transfer Learning.

Cardiac or Infectious? Transfer Learning with Chest X-Rays for ER Patient Classification

One of the challenges with urgent evaluation of patients with acute respiratory distress syndrome (ARDS) in the emergency room (ER) is distinguishing between cardiac vs infectious etiologies for their pulmonary findings. We evaluated ER patient classification for cardiac and infection causes with clinical data and chest X-ray image data. We show that a deep-learning model trained with an external image data set can be used to extract image features and improve the classification accuracy of a data set that does not contain enough image data to train a deep-learning model. We also conducted clinical feature importance analysis and identified the most important clinical features for ER patient classification. This model can be upgraded to include a SARS-CoV-2 specific classification with COVID-19 patients data. The current model is publicly available with an interface at the web link: http://nbttranslationalresearch.org/. Data statementThe clinical data and chest x-ray image data for this study were collected and prepared by the residents and researchers of the Joint Translational Research Lab of Arkansas State University (A-State) and St. Bernards Medical Center (SBMC) Internal Medicine Residency Program. As data collection is on-going for the project stage-II of clinical testing, raw data is not currently available for data sharing to the public. EthicsThis study was approved by the St. Bernards Medical Centers Institutional Review Board (IRB).

Improvement and application of the opening method of mechanical ventilator humidification tank liquid / 中国实用护理杂志

Objective@#To improve the opening method of humidifier, enhance the efficiency and success rate of opening the humidifiers.@*Methods@#Use all the recycled humidifier from respirators in the wards from 1st of May to 31st of October 2015 as control group. Use the recycled humidifiers from 1st Feb to 31st July 2016 as the test group. Use metal tools to open humidifier bottles in the control group while use the newly developed and improved opening device to open the bottles in the test group to test and compare the damage rate.@*Results@#The testing group that used the practical opening device for humidifier has a decrease in damage rate from 5.97% (16/268) to 1.22% (3/245), and this difference before and after was statistically significant (χ2=8.082, P<0.01).@*Conclusions@#The use of the opening device for humidifier can enhance the intact rate, greatly improve working efficiency and save the cost.

Investigation of Species of Wild Chinese Medicinal Herbs for Treating Liver Diseases in Quanzhou City of Fujian Province / 中国中医药信息杂志

ObjectiveTo investigate the species of wild Chinese medicinal herbs for treating liver diseases in Quanzhou City of Fujian Province.Methods Based on the field investigation combined with literature reference, induction and analysis of the species of wild Chinese medicinal herbs for treating liver diseases in Quanzhou City of Fujian Province have been conducted.ResultsIt was found that there are 59 families, 112 genera, and 132 species of wild Chinese medicinal herbs for treating liver diseases in Quanzhou City of Fujian Province, most of which belong to angiosperm. 15 families, 21 genera, and 21 species are recorded in the Chinese Pharmacopoeia (2010 edition). Compositae has the largest number of genera and species (16 genera &19 species), followed by leguminosae (8 genera & 8 species) and labiatae (6 genera & 8 species). Chinese medicinal herbs for heat-clearing and promoting urination and leaching out damp amount to 69 and 20 species respectively. Over 88% of these wild medicinal herbs can be used to treat various liver diseases, and 46 species are used with high frequency. They are mainly taken by water decoction and oral administration.ConclusionWild Chinese medicinal herbs for treating liver diseases in Quanzhou City of Fujian Province are rich in species and resources, and are worth further development and utilization.

Correlation between exercise lowering blood pressure and arterial baroreflex sensitivity in spontaneously hypertensive rats / 心血管康复医学杂志

Objectives: To explore the influence of exercise training on the arterial baroreflex sensitivity （BRS）and correlation between blood pressure and BRS in spontaneously hypertensive rats (SHR). Methods: Male SHR（n=20）and normotensive Wistar rats（n=20）were randomly assigned to normality group and exercise group, n=10 in each group. Rats in two exercise groups received treadmill training at a speed of 20 m/min for 60 min/d, 6 d/w for eight weeks. Systolic blood pressure (SBP) and heart rate (HR) were measured using a tail-cuff method in a conscious state. Intravenous injections of phenylephrine (PE) and sodium nitroprusside (NP) were used to induce depressor and pressor reflex respectively. The ratio of HR over mean arterial pressure (MAP) (HR /MAP) after administration of PE or NP was regarded as an index of depressor reflex sensitivity (BRS-PE) and pressor reflex sensivity (BRS-NP). Results: After eight-week exercise training, compared with SHR normality group, there were significant reduction in resting SBP [(180±8.5) mmHg vs. (163.6±10.7) mmHg] and in HR [(368.4±13.3) beats/min vs. (345.0±9.8) beats/min] in SHR exercise group, P0.05), compared with Wistar normality group, there was significant reduction in HR [(362.2 ± 13.0) beats/min vs. (343.9 ± 10.2) beats/min, P <0.05] in Wistar exercise group. Compared with SHR normality group, there were significant rise in BRS [BRS-PE: (0.89 ± 0.13) bpm/mmHg vs. (1.32 ± 0.22) bpm/mmHg, BRS-NP： (0.60± 0.09) bpm/mmHg vs. (1.21± 0.26) bpm/mmHg, P<0.01] in SHR exercise group, but still lower than those of Wistar normality group [BRS-PE: (1.96±0.23) bpm/mmHg, BRS-NP: (1.32±0.17) bpm/mmHg]. Pearson linear correlation analysis indicated that MAP was significantly inversely correlated with BRS (r=-0.734, P<0.01) in SHR normality and exercise group. Conclusion: Exercise training may significantly decrease SHR blood pressure; it is related to improved baroreflex sensitivity induced by exercise, indicating that enhanced baroreflex may be an important mechanism of exercise therapy in hypertensive patients.

The study on the expression pattern of MMP-2 and E-Cadherin in the effects of gastrin on gastric cancer cells / 重庆医学

Objective To approach on the expression pattern of MMP-2 and E-Cadherin in the effects of gastrin on gastric canc-er cells ,so as to investigate their role in the metastasis and invasion of gastric cancer .Methods Genomic DNA modified by sodium bisulfite was used as a template .RQ-MSP detect the methylation status of the gene promoter .The relative amount of Methylation status was performed with comparative threshold cycle (2- △ t ) method .Results MMP-2 and E-Cadherin genes in gastric cancer cells were at hemimethylated status in gastrin before and after the intervention .Gastrin can reduce the methylation levels of MMP-2 gene(P0 .05) .Gastrin can pro-mote the methylation levels of MMP-2 gene ,but proglumide decrease that effect (P<0 .05) .Conclusion Gastrin plays a certain role in the biological behavior of gastric cancer impact by changeing on the expression pattern of MMP-2 and E-Cadherin in the effects of gastrin on gastric cancer cells .