COVID-19 Detection from Cough Recordings Using Bag-of-Words Classifiers.

Reliable detection of COVID-19 from cough recordings is evaluated using bag-of-words classifiers. The effect of using four distinct feature extraction procedures and four different encoding strategies is evaluated in terms of the Area Under Curve (AUC), accuracy, sensitivity, and F1-score. Additional studies include assessing the effect of both input and output fusion approaches and a comparative analysis against 2D solutions using Convolutional Neural Networks. Extensive experiments conducted on the COUGHVID and COVID-19 Sounds datasets indicate that sparse encoding yields the best performances, showing robustness against various combinations of feature type, encoding strategy, and codebook dimension parameters.

IoMT Meets Machine Learning: From Edge to Cloud Chronic Diseases Diagnosis System.

In conventional healthcare, real-time monitoring of patient records and information mining for timely diagnosis of chronic diseases under certain health conditions is a crucial process. Chronic diseases, if not diagnosed in time, may result in patients' death. In modern medical and healthcare systems, Internet of Things (IoT) driven ecosystems use autonomous sensors to sense and track patients' medical conditions and suggest appropriate actions. In this paper, a novel IoT and machine learning (ML)-based hybrid approach is proposed that considers multiple perspectives for early detection and monitoring of 6 different chronic diseases such as COVID-19, pneumonia, diabetes, heart disease, brain tumor, and Alzheimer's. The results from multiple ML models are compared for accuracy, precision, recall, F1 score, and area under the curve (AUC) as a performance measure. The proposed approach is validated in the cloud-based environment using benchmark and real-world datasets. The statistical analyses on the datasets using ANOVA tests show that the accuracy results of different classifiers are significantly different. This will help the healthcare sector and doctors in the early diagnosis of chronic diseases.

Circulating mid-regional proadrenomedullin is a predictor of mortality in patients with COVID-19: a systematic review and meta-analysis.

BACKGROUND: Although there is increasing understanding of the changes in the laboratory parameters of Coronavirus disease 2019 (COVID-19), the correlation between circulating Mid-regional Proadrenomedullin (MR-proADM) and mortality of patients with COVID-19 is not fully understood. In this study, we conducted a systematic review and meta-analysis to evaluate the prognostic value of MR-proADM in patients with COVID-19. METHODS: The PubMed, Embase, Web of Science, Cochrane Library, Wanfang, SinoMed and Chinese National Knowledge Infrastructure (CNKI) databases were searched from 1 January 2020 to 20 March 2022 for relevant literature. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) was used to assess quality bias, STATA was employed to pool the effect size by a random effects model, and potential publication bias and sensitivity analyses were performed. RESULTS: 14 studies comprising 1822 patients with COVID-19 met the inclusion criteria, there were 1145 (62.8%) males and 677 (31.2%) females, and the mean age was 63.8 ± 16.1 years. The concentration of MR-proADM was compared between the survivors and non-survivors in 9 studies and the difference was significant (P < 0.01), I2 = 46%. The combined sensitivity was 0.86 [0.73-0.92], and the combined specificity was 0.78 [0.68-0.86]. We drew the summary receiver operating characteristic (SROC) curve and calculated the area under curve (AUC) = 0.90 [0.87-0.92]. An increase of 1 nmol/L of MR-proADM was independently associated with a more than threefold increase in mortality (odds ratio (OR) 3.03, 95% confidence interval (CI) 2.26-4.06, I2 = 0.0%, P = 0.633). The predictive value of MR-proADM for mortality was better than many other biomarkers. CONCLUSION: MR-proADM had a very good predictive value for the poor prognosis of COVID-19 patients. Increased levels of MR-proADM were independently associated with mortality in COVID-19 patients and may allow a better risk stratification.

Evaluation of the Drug-Drug Interaction Potential of Ensitrelvir Fumaric Acid with Cytochrome P450 3A Substrates in Healthy Japanese Adults.

BACKGROUND: Management of drug-drug interactions (DDIs) for ensitrelvir, a novel 3-chymotrypsin-like protease inhibitor of SARS-CoV-2 infection is crucial. A previous clinical DDI study of ensitrelvir with midazolam, a clinical index cytochrome P450 (CYP) 3A substrate, demonstrated that ensitrelvir given for 5 days orally with a loading/maintenance dose of 750/250 mg acted as a strong CYP3A inhibitor. OBJECTIVES: The objectives of this study were to investigate the effect of ensitrelvir on the pharmacokinetics of CYP3A substrates, dexamethasone, prednisolone and midazolam, and to assess the pharmacokinetics, safety, and tolerability of ensitrelvir following multiple-dose administration of ensitrelvir. METHODS: This was a Phase 1, multicenter, single-arm, open-label study in healthy Japanese adult participants. The effects of multiple doses of ensitrelvir in the fasted state on the pharmacokinetics of dexamethasone, prednisolone, and midazolam were investigated. Ensitrelvir was administered from Day 1 through Day 5, with a loading/maintenance dose of 750/250 mg for the dexamethasone and prednisolone cohorts whereas 375/125 mg for the midazolam cohort. Either dexamethasone, prednisolone, or midazolam was administered alone (Day - 2) or in combination with ensitrelvir (Day 5) in each of the cohorts. Additionally, dexamethasone or prednisolone was administered on Days 9 and 14. The pharmacokinetic parameters of ensitrelvir, dexamethasone, prednisolone, and midazolam were calculated based on their plasma concentration data with non-compartmental analysis. In safety assessments, the nature, frequency, and severity of treatment-emergent adverse events were evaluated and recorded. RESULTS: The area under the concentration-time curve (AUC) ratio of dexamethasone on Day 5 was 3.47-fold compared with the corresponding values for dexamethasone alone on Day - 2 and the effect diminished over time after the last dose of ensitrelvir. No clinically meaningful effect was observed for prednisolone. The AUC ratio of midazolam was 6.77-fold with ensitrelvir 375/125 mg suggesting ensitrelvir at 375/125 mg strongly inhibits CYP3A similar to that at 750/250 mg. No new safety signals with ensitrelvir were reported during the study. CONCLUSION: The inhibitory effect for CYP3A was confirmed after the last dose of ensitrelvir, and the effect diminished over time. In addition, ensitrelvir at 375/125 mg showed CYP3A inhibitory potential similar to that at 750/250 mg. These findings can be used as a clinical recommendation for prescribing ensitrelvir with regard to concomitant medications. CLINICAL TRIAL REGISTRATION: Japan Registry of Clinical Trials identifier: jRCT2031210202.

Results From a Randomized, Open-Label, Crossover Study Evaluating the Effect of the Aldosterone Synthase Inhibitor Baxdrostat on the Pharmacokinetics of Metformin in Healthy Human Subjects.

BACKGROUND: Due to the high comorbidity of diabetes and hypertension, co-administration of metformin with anti-hypertensive drugs is likely. Baxdrostat is an aldosterone synthase inhibitor in development for the potential treatment of hypertension. In vitro data indicated that baxdrostat inhibits the multidrug and toxin extrusion 1 (MATE1) and MATE2-K renal transporters. Metformin is a MATE substrate, so this study assessed potential effects of baxdrostat on the pharmacokinetics of metformin. METHODS: Twenty-seven healthy volunteers received 1000 mg metformin alone and 1000 mg metformin in the presence of 10 mg baxdrostat in a randomized, crossover manner. Each treatment was separated by 10 or more days. Blood and urine samples were collected over a 3-day period after each treatment to measure plasma and urine concentrations of metformin. Safety was assessed by adverse events (AEs), physical examinations, electrocardiograms, vital signs, and clinical laboratory evaluations. RESULTS: There were no deaths, serious AEs, discontinuations due to treatment-emergent AEs, or noteworthy increases in AEs with either treatment, indicating that metformin and baxdrostat were well-tolerated when co-administered. Baxdrostat did not significantly affect plasma concentrations or renal clearance of metformin. CONCLUSION: The results of this study suggest that diabetic patients with hypertension receiving both metformin and baxdrostat are unlikely to require dose adjustment. REGISTRATION: ClinicalTrials.gov identifier no. NCT05526690.

A Real Time Method for Distinguishing COVID-19 Utilizing 2D-CNN and Transfer Learning.

Rapid identification of COVID-19 can assist in making decisions for effective treatment and epidemic prevention. The PCR-based test is expert-dependent, is time-consuming, and has limited sensitivity. By inspecting Chest R-ray (CXR) images, COVID-19, pneumonia, and other lung infections can be detected in real time. The current, state-of-the-art literature suggests that deep learning (DL) is highly advantageous in automatic disease classification utilizing the CXR images. The goal of this study is to develop models by employing DL models for identifying COVID-19 and other lung disorders more efficiently. For this study, a dataset of 18,564 CXR images with seven disease categories was created from multiple publicly available sources. Four DL architectures including the proposed CNN model and pretrained VGG-16, VGG-19, and Inception-v3 models were applied to identify healthy and six lung diseases (fibrosis, lung opacity, viral pneumonia, bacterial pneumonia, COVID-19, and tuberculosis). Accuracy, precision, recall, f1 score, area under the curve (AUC), and testing time were used to evaluate the performance of these four models. The results demonstrated that the proposed CNN model outperformed all other DL models employed for a seven-class classification with an accuracy of 93.15% and average values for precision, recall, f1-score, and AUC of 0.9343, 0.9443, 0.9386, and 0.9939. The CNN model equally performed well when other multiclass classifications including normal and COVID-19 as the common classes were considered, yielding accuracy values of 98%, 97.49%, 97.81%, 96%, and 96.75% for two, three, four, five, and six classes, respectively. The proposed model can also identify COVID-19 with shorter training and testing times compared to other transfer learning models.

Effect of Various Dosing Schedules on the Pharmacokinetics of Oral Semaglutide: A Randomised Trial in Healthy Subjects.

BACKGROUND: Prescribing information instructs taking oral semaglutide (a glucagon-like peptide-1 analogue) in the fasting state, followed by a post-dose fasting period of ≥ 30 min. This trial compared the recommended dosing schedule with alternative schedules. METHODS: This was a randomised, single-centre, multiple-dose, open-label, five-armed, parallel-group trial in healthy subjects who received once-daily oral semaglutide (3 mg for 5 days followed by 7 mg for 5 days). Subjects (n = 156) were randomised to five dosing schedules: 2-, 4-, or 6-h pre-dose fast followed by a 30-min post-dose fast (treatment arms: 2 h-30 min, 4-30 min, 6 h-30 min); 2-h pre-dose fast followed by an overnight post-dose fast (treatment arm: 2 h-night); or overnight pre-dose fast followed by a 30-min post-dose fast (reference arm: night-30 min). Semaglutide plasma concentration was measured regularly until 24 h after the 10th dose. Endpoints included area under the semaglutide plasma concentration-time curve during a 24-h interval after the 10th dose (AUC0-24h) (primary endpoint) and maximum observed semaglutide plasma concentration after the 10th dose (Cmax) (secondary endpoint). RESULTS: Compared with an overnight pre-dose fast (reference arm: night-30 min), shorter pre-dose fasting times in the 2 h-night, 2 h-30 min, 4 h-30 min, and 6 h-30 min treatment arms resulted in significantly lower semaglutide AUC0-24h and Cmax after the 10th dose (estimated treatment ratio ranges: 0.12-0.43 and 0.11-0.44, respectively; p < 0.0001 for all comparisons). Semaglutide AUC0-24h and Cmax after the 10th dose were similar for the 2 h-30 min and 2 h-night treatment arms. CONCLUSION: This trial supports dosing oral semaglutide in accordance with prescribing information, which requires dosing in the fasting state. TRIAL REGISTRATION: ClinicalTrials.gov (NCT04513704); registered August 14, 2020.

Oral semaglutide is a human glucagon-like peptide-1 analogue that has been approved for the treatment of type 2 diabetes. It has been established that taking oral semaglutide with food or large volumes of water decreases absorption of the drug in the body. Current prescribing information instructs taking oral semaglutide on an empty stomach (known as the fasting state), with 120 mL/4 oz of water, then waiting for at least 30 min before consuming any food, water, or taking other oral medications. This study investigates whether different dosing schedules for oral semaglutide could potentially offer more flexibility to patients in the timing of their oral semaglutide dosing. The trial, conducted in healthy volunteers, compares the dosing schedule described in the prescribing information with different fasting times before (pre-dose) and after (post-dose) taking oral semaglutide during the day or evening, to see if there were any effects on the concentration of drug in the body. Compared to the recommended overnight fasting period, shorter pre-dose fasting periods of 2­6 h with a 30-min post-dose fast considerably reduced semaglutide exposure in the body. Similarly, semaglutide exposure was also reduced with a 2-h pre-dose fast combined with post-dose overnight fasting. These findings further support the current prescribing information, which states that patients should take their oral semaglutide dose after an overnight fast.

Does the Ethnic Difference Affect the Pharmacokinetics of Favipiravir? A Pharmacokinetic Study in Healthy Egyptian Volunteers and Development of Level C In-vitro In-vivo Correlation.

Favipiravir is an antiviral drug used to treat influenza and is also being investigated for the treatment of SARS-CoV-2. Its pharmacokinetic profile varies depending on ethnic group. The present research examines the pharmacokinetic features of favipiravir in healthy male Egyptian volunteers. Another goal of this research is to determine the optimum dissolution testing conditions for immediate release tablets. In vitro dissolution testing was investigated for favipiravir tablets in three different pH media. The pharmacokinetic features of favipiravir were examined in 27 healthy male Egyptian volunteers. The parameter "AUC0-t" vs. percent dissolved was used to develop level C in vitro in vivo correlation (IVIVC) to set the optimum dissolution medium to achieve accurate dissolution profile for favipiravir (IR) tablets. The in vitro release results revealed significant difference among the three different dissolution media. The Pk parameters of twenty-seven human subjects showed mean value of Cpmax of 5966.45 ng/mL at median tmax of 0.75 h with AUC0-∞ equals 13325.54 ng.h/mL, showing half-life of 1.25 h. Level C IVIVC was developed successfully. It was concluded that Egyptian volunteers had comparable Pk values to American and Caucasian volunteers, however they were considerably different from Japanese subjects. AUC0-t vs. % dissolved was used to develop level C IVIVC to set the optimum dissolution medium. Phosphate buffer medium (pH 6.8) was found to be the optimum dissolution medium for in vitro dissolution testing for Favipiravir IR tablets.

Fibrosis-4 index and mortality in coronavirus disease 2019: a meta-analysis.

BACKGROUND/AIMS: In this meta-analysis, we aimed to evaluate the prognostic value of fibrosis-4 index (FIB-4) in COVID-19. METHODS: We performed a comprehensive literature search of PubMed, Embase, and Scopus databases on 26 November 2020. FIB-4 was calculated by [age (years) × AST (IU/L)]/[platelet count (109/L) × âˆšALT (U/L)]. A value above cutoff point was considered high and a value below cutoff point was considered low. The main outcome was mortality, the association between high FIB-4 and mortality was reported in odds ratio (OR). Sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic OR (DOR), area under the curve (AUC) were generated. RESULTS: There were 963 patients from five studies included in this systematic review and meta-analysis. Meta-analysis showed that high FIB-4 was associated with increased mortality [OR 3.96 (2.16-7.27), P < 0.001; I2: 41.3%]. High FIB-4 was associated mortality with a sensitivity of 0.56 (0.40-0.70), specificity of 0.80 (0.72-0.86), PLR 2.8 (1.8-4.2), NLR 0.55 (0.39-0.78), DOR 5 (2-10), and AUC of 0.77 (0.73-0.81). Fagan's nomogram indicated that for a pre-test probability (mortality) of 30%, a high FIB-4 was associated with 54% post-test probability and a low FIB-4 was associated with 19%, respectively. The funnel-plot analysis was asymmetrical, trim-and-fill analysis by imputation of a study on the left side using linear estimator resulted in an OR of 3.48 (1.97-6.14). Egger's test showed no indication of small-study effects (P = 0.881). CONCLUSION: High FIB-4 was associated with mortality in patients with COVID-19.

IRCM-Caps: An X-ray image detection method for COVID-19.

OBJECTIVE: COVID-19 is ravaging the world, but traditional reverse transcription-polymerase reaction (RT-PCR) tests are time-consuming and have a high false-negative rate and lack of medical equipment. Therefore, lung imaging screening methods are proposed to diagnose COVID-19 due to its fast test speed. Currently, the commonly used convolutional neural network (CNN) model requires a large number of datasets, and the accuracy of the basic capsule network for multiple classification is limital. For this reason, this paper proposes a novel model based on CNN and CapsNet. METHODS: The proposed model integrates CNN and CapsNet. And attention mechanism module and multi-branch lightweight module are applied to enhance performance. Use the contrast adaptive histogram equalization (CLAHE) algorithm to preprocess the image to enhance image contrast. The preprocessed images are input into the network for training, and ReLU was used as the activation function to adjust the parameters to achieve the optimal. RESULT: The test dataset includes 1200 X-ray images (400 COVID-19, 400 viral pneumonia, and 400 normal), and we replace CNN of VGG16, InceptionV3, Xception, Inception-Resnet-v2, ResNet50, DenseNet121, and MoblieNetV2 and integrate with CapsNet. Compared with CapsNet, this network improves 6.96%, 7.83%, 9.37%, 10.47%, and 10.38% in accuracy, area under the curve (AUC), recall, and F1 scores, respectively. In the binary classification experiment, compared with CapsNet, the accuracy, AUC, accuracy, recall rate, and F1 score were increased by 5.33%, 5.34%, 2.88%, 8.00%, and 5.56%, respectively. CONCLUSION: The proposed embedded the advantages of traditional convolutional neural network and capsule network and has a good classification effect on small COVID-19 X-ray image dataset.

Leveraging Prior Healthy Participant Pharmacokinetic Data to Evaluate the Impact of Renal and Hepatic Impairment on Ritlecitinib Pharmacokinetics.

Ritlecitinib is a selective, covalent, irreversible inhibitor of Janus kinase 3 (JAK3) and the tyrosine kinase expressed in hepatocellular carcinoma (TEC) family kinases. Pharmacokinetics and safety of ritlecitinib in participants with hepatic (Study 1) or renal (Study 2) impairment were to be characterized from two phase I studies. Due to a study pause caused by the COVID-19 pandemic, the study 2 healthy participant (HP) cohort was not recruited; however, the demography of the severe renal impairment cohort closely matched the study 1 HP cohort. We present results from each study and two innovative approaches to utilizing available HP data as reference data for study 2: a statistical approach using analysis of variance and an in silico simulation of an HP cohort created using a population pharmacokinetics (POPPK) model derived from several ritlecitinib studies. For study 1, the observed area under the curve for 24-h dosing interval and maximum plasma concentration for HPs and their observed geometric mean ratios (participants with moderate hepatic impairment vs HPs) were within 90% prediction intervals from the POPPK simulation-based approach, thereby validating the latter approach. When applied to study 2, both the statistical and POPPK simulation approaches demonstrated that patients with renal impairment would not require ritlecitinib dose modification. In both phase I studies, ritlecitinib was generally safe and well tolerated. These analyses represent a new methodology for generating reference HP cohorts in special population studies for drugs in development with well-characterized pharmacokinetics in HPs and adequate POPPK models. TRIAL REGISTRATION: ClinicalTrials.gov NCT04037865 , NCT04016077 , NCT02309827 , NCT02684760 , and NCT02969044 .

Clinical assessment of SNIBE Maglumi SARS-CoV-2 antigen fully-automated chemiluminescent immunoassay.

OBJECTIVES: Given that SARS-CoV-2 antigen tests will represent a pillar for supporting or surrogating molecular testing in the endemic period, we report here the clinical performance of the new SNIBE Maglumi SARS-CoV-2 antigen fully-automated chemiluminescent immunoassay (MAG-CLIA SARS-CoV-2 Ag). METHODS: The study population consisted of 181 subjects (mean age 61 ± 21 years; 92 females) undergoing coronavirus disease 2019 (COVID-19) testing at the local diagnostic facility, from December 2022 to February 2023. Routine diagnostic practice involved the collection of a double nostril nasopharyngeal swab, analyzed in duplicate with SARS-CoV-2 antigen (MAG-CLIA SARS-CoV-2 Ag) and molecular (Altona Diagnostics RealStar SARS-CoV-2 RT-PCR Kit) tests. RESULTS: A significant Spearman's correlation was found between MAG-CLIA SARS-CoV-2 Ag and mean Ct values of SARS-CoV-2 E and S genes (r=-0.95; p<0.001). In all nasopharyngeal samples, the area under the curve (AUC) of MAG-CLIA SARS-CoV-2 Ag was 0.86 (95% CI, 0.81-0.90), with 0.71 sensitivity and 1.00 specificity at 7 ng/L cut-off, increasing to 0.98 (95% CI, 0.96-1.00) AUC and 0.96 sensitivity (with 0.97 specificity) in high viral load samples. When SARS-CoV-2 N protein concentration was replaced with raw instrumental readings (i.e., relative light units [RLU]), the AUC in all samples increased to 0.94. A RLU value of 945 was associated with 88.4% accuracy, 0.85 sensitivity, 0.95 specificity, 0.77 negative predictive value (NPV) and 0.97 positive predictive value (PPV), respectively. CONCLUSIONS: We found satisfactory analytical performance of MAG-CLIA SARS-CoV-2 Ag, which could be used as surrogate of molecular testing for identifying high viral load samples. Broadening the reportable range of values may generate even better performance.

Machine Learning Models Using Routinely Collected Clinical Data Offer Robust and Interpretable Predictions of 90-Day Unplanned Acute Care Use for Cancer Immunotherapy Patients.

PURPOSE: Clinical management of patients receiving immune checkpoint inhibitors (ICIs) could be informed using accurate predictive tools to identify patients at risk of short-term acute care utilization (ACU). We used routinely collected data to develop and assess machine learning (ML) algorithms to predict unplanned ACU within 90 days of ICI treatment initiation. METHODS: We used aggregated electronic health record data from 7,960 patients receiving ICI treatments to train and assess eight ML algorithms. We developed the models using pre-SARS-COV-19 COVID-19 data generated between January 2016 and February 2020. We validated our algorithms using data collected between March 2020 and June 2022 (peri-COVID-19 sample). We assessed performance using area under the receiver operating characteristic curves (AUROC), sensitivity, specificity, and calibration plots. We derived intuitive explanations of predictions using variable importance and Shapley additive explanation analyses. We assessed the marginal performance of ML models compared with that of univariate and multivariate logistic regression (LR) models. RESULTS: Most algorithms significantly outperformed the univariate and multivariate LR models. The extreme gradient boosting trees (XGBT) algorithm demonstrated the best overall performance (AUROC, 0.70; sensitivity, 0.53; specificity, 0.74) on the peri-COVID-19 sample. The algorithm performance was stable across both pre- and peri-COVID-19 samples, as well as ICI regimen and cancer groups. Type of ICI agents, oxygen saturation, diastolic blood pressure, albumin level, platelet count, immature granulocytes, absolute monocyte, chloride level, red cell distribution width, and alcohol intake were the top 10 key predictors used by the XGBT algorithm. CONCLUSION: Machine learning algorithms trained using routinely collected data outperformed traditional statistical models when predicting 90-day ACU. The XGBT algorithm has the potential to identify high-ACU risk patients and enable preventive interventions to avoid ACU.

Multi-domain medical image translation generation for lung image classification based on generative adversarial networks.

OBJECTIVE: Lung image classification-assisted diagnosis has a large application market. Aiming at the problems of poor attention to existing translation models, the insufficient ability of key transfer and generation, insufficient quality of generated images, and lack of detailed features, this paper conducts research on lung medical image translation and lung image classification based on generative adversarial networks. METHODS: This paper proposes a medical image multi-domain translation algorithm MI-GAN based on the key migration branch. After the actual analysis of the imbalanced medical image data, the key target domain images are selected, the key migration branch is established, and a single generator is used to complete the medical image multi-domain translation. The conversion between domains ensures the attention performance of the medical image multi-domain translation model and the quality of the synthesized images. At the same time, a lung image classification model based on synthetic image data augmentation is proposed. The synthetic lung CT medical images and the original real medical images are used as the training set together to study the performance of the auxiliary diagnosis model in the classification of normal healthy subjects, and also of the mild and severe COVID-19 patients. RESULTS: Based on the chest CT image dataset, MI-GAN has completed the mutual conversion and generation of normal lung images without disease, viral pneumonia and Mild COVID-19 images. The synthetic images GAN-test and GAN-train indicators reached, respectively 92.188% and 85.069%, compared with other generative models in terms of authenticity and diversity, there is a considerable improvement. The accuracy rate of pneumonia diagnosis of the lung image classification model is 93.85%, which is 3.1% higher than that of the diagnosis model trained only with real images; the sensitivity of disease diagnosis is 96.69%, a relative improvement of 7.1%. 1%, the specificity was 89.70%; the area under the ROC curve (AUC) increased from 94.00% to 96.17%. CONCLUSION: In this paper, a multi-domain translation model of medical images based on the key transfer branch is proposed, which enables the translation network to have key transfer and attention performance. It is verified on lung CT images and achieved good results. The required medical images are synthesized by the above medical image translation model, and the effectiveness of the synthesized images on the lung image classification network is verified experimentally.

Stratification of COVID-19 Severity Using SeptiCyte RAPID, a Novel Host Immune Response Test.

SeptiCyte® RAPID is a gene expression assay measuring the relative expression levels of host response genes PLA2G7 and PLAC8, indicative of a dysregulated immune response during sepsis. As severe forms of COVID-19 may be considered viral sepsis, we evaluated SeptiCyte RAPID in a series of 94 patients admitted to Foch Hospital (Suresnes, France) with proven SARS-CoV-2 infection. EDTA blood was collected in the emergency department (ED) in 67 cases, in the intensive care unit (ICU) in 23 cases and in conventional units in 4 cases. SeptiScore (0-15 scale) increased with COVID-19 severity. Patients in ICU had the highest SeptiScores, producing values comparable to 8 patients with culture-confirmed bacterial sepsis. Receiver operating characteristic (ROC) curve analysis had an area under the curve (AUC) of 0.81 for discriminating patients requiring ICU admission from patients who were immediately discharged or from patients requiring hospitalization in conventional units. SeptiScores increased with the extent of the lung injury. For 68 patients, a chest computed tomography (CT) scan was performed within 24 h of COVID-19 diagnosis. SeptiScore >7 suggested lung injury ≥50% (AUC = 0.86). SeptiCyte RAPID was compared to other biomarkers for discriminating Critical + Severe COVID-19 in ICU, versus Moderate + Mild COVID-19 not in ICU. The mean AUC for SeptiCyte RAPID was superior to that of any individual biomarker or combination thereof. In contrast to C-reactive protein (CRP), correlation of SeptiScore with lung injury was not impacted by treatment with anti-inflammatory agents. SeptiCyte RAPID can be a useful tool to identify patients with severe forms of COVID-19 in ED, as well as during follow-up.

Influence of social deprivation index on in-hospital outcomes of COVID-19.

While it is known that social deprivation index (SDI) plays an important role on risk for acquiring Coronavirus Disease 2019 (COVID-19), the impact of SDI on in-hospital outcomes such as intubation and mortality are less well-characterized. We analyzed electronic health record data of adults hospitalized with confirmed COVID-19 between March 1, 2020 and February 8, 2021 from the INSIGHT Clinical Research Network (CRN). To compute the SDI (exposure variable), we linked clinical data using patient's residential zip-code with social data at zip-code tabulation area. SDI is a composite of seven socioeconomic characteristics determinants at the zip-code level. For this analysis, we categorized SDI into quintiles. The two outcomes of interest were in-hospital intubation and mortality. For each outcome, we examined logistic regression and random forests to determine incremental value of SDI in predicting outcomes. We studied 30,016 included COVID-19 patients. In a logistic regression model for intubation, a model including demographics, comorbidity, and vitals had an Area under the receiver operating characteristic curve (AUROC) = 0.73 (95% CI 0.70-0.75); the addition of SDI did not improve prediction [AUROC = 0.73 (95% CI 0.71-0.75)]. In a logistic regression model for in-hospital mortality, demographics, comorbidity, and vitals had an AUROC = 0.80 (95% CI 0.79-0.82); the addition of SDI in Model 2 did not improve prediction [AUROC = 0.81 (95% CI 0.79-0.82)]. Random forests revealed similar findings. SDI did not provide incremental improvement in predicting in-hospital intubation or mortality. SDI plays an important role on who acquires COVID-19 and its severity; but once hospitalized, SDI appears less important.

Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran.

BACKGROUND: The high number of COVID-19 deaths is a serious threat to the world. Demographic and clinical biomarkers are significantly associated with the mortality risk of this disease. This study aimed to implement Generalized Neural Additive Model (GNAM) as an interpretable machine learning method to predict the COVID-19 mortality of patients. METHODS: This cohort study included 2181 COVID-19 patients admitted from February 2020 to July 2021 in Sina and Besat hospitals in Hamadan, west of Iran. A total of 22 baseline features including patients' demographic information and clinical biomarkers were collected. Four strategies including removing missing values, mean, K-Nearest Neighbor (KNN), and Multivariate Imputation by Chained Equations (MICE) imputation methods were used to deal with missing data. Firstly, the important features for predicting binary outcome (1: death, 0: recovery) were selected using the Random Forest (RF) method. Also, synthetic minority over-sampling technique (SMOTE) method was used for handling imbalanced data. Next, considering the selected features, the predictive performance of GNAM for predicting mortality outcome was compared with logistic regression, RF, generalized additive model (GAMs), gradient boosting decision tree (GBDT), and deep neural networks (DNNs) classification models. Each model trained on fifty different subsets of a train-test dataset to ensure a model performance. The average accuracy, F1-score and area under the curve (AUC) evaluation indices were used for comparison of the predictive performance of the models. RESULTS: Out of the 2181 COVID-19 patients, 624 died during hospitalization and 1557 recovered. The missing rate was 3 percent for each patient. The mean age of dead patients (71.17 ± 14.44 years) was statistically significant higher than recovered patients (58.25 ± 16.52 years). Based on RF, 10 features with the highest relative importance were selected as the best influential features; including blood urea nitrogen (BUN), lymphocytes (Lym), age, blood sugar (BS), serum glutamic-oxaloacetic transaminase (SGOT), monocytes (Mono), blood creatinine (CR), neutrophils (NUT), alkaline phosphatase (ALP) and hematocrit (HCT). The results of predictive performance comparisons showed GNAM with the mean accuracy, F1-score, and mean AUC in the test dataset of 0.847, 0.691, and 0.774, respectively, had the best performance. The smooth function graphs learned from the GNAM were descending for the Lym and ascending for the other important features. CONCLUSIONS: Interpretable GNAM can perform well in predicting the mortality of COVID-19 patients. Therefore, the use of such a reliable model can help physicians to prioritize some important demographic and clinical biomarkers by identifying the effective features and the type of predictive trend in disease progression.