COVID-19 Diagnosis on Chest Radiograph Using Artificial Intelligence.

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic has disrupted the world since 2019, causing significant morbidity and mortality in developed and developing countries alike. Although substantial resources have been diverted to developing diagnostic, preventative, and treatment measures, disparities in the availability and efficacy of these tools vary across countries. We seek to assess the ability of commercial artificial intelligence (AI) technology to diagnose COVID-19 by analyzing chest radiographs. MATERIALS AND METHODS: Chest radiographs taken from symptomatic patients within two days of polymerase chain reaction (PCR) tests were assessed for COVID-19 infection by board-certified radiologists and commercially available AI software. Sixty patients with negative and 60 with positive COVID reverse transcription-polymerase chain reaction (RT-PCR) tests were chosen. Results were compared against results of the PCR test for accuracy and statistically analyzed by receiver operating characteristic (ROC) curves along with area under the curve (AUC) values. RESULTS: A total of 120 chest radiographs (60 positive and 60 negative RT-PCR tests) radiographs were analyzed. The AI software performed significantly better than chance (p = 0.001) and did not differ significantly from the radiologist ROC curve (p = 0.78). CONCLUSION: Commercially available AI software was not inferior compared with trained radiologists in accurately identifying COVID-19 cases by analyzing radiographs. While RT-PCR testing remains the standard, current advances in AI help correctly analyze chest radiographs to diagnose COVID-19 infection.

Determining the SARS-CoV-2 Anti-Spike Cutoff Level Denoting Neutralizing Activity Using Two Commercial Kits.

BACKGROUND: The viral neutralization assay is the gold standard to estimate the level of immunity against SARS-CoV-2. This study analyzes the correlation between the quantitative Anti-SARS-CoV-2 QuantiVac ELISA (IgG) and the NeutraLISA neutralization assay. METHODS: 650 serum samples were tested for both SARS-CoV-2 anti-spike (anti-S) immunoglobulin G (IgG) and neutralizing antibodies (nAbs) using kits by EUROIMMUN, Germany. RESULTS: There was a significant correlation between levels of anti-S and nAbs (Spearman's rho = 0.913). Among the positive samples for anti-S, 77.0% (n = 345) were positive for nAbs. There was a substantial agreement between anti-S and nAbs (Cohen's kappa coefficient = 0.658; agreement of 83.38%). Considering NeutraLISA as a gold standard, anti-S had a sensitivity of 98.57%, specificity of 65.66%, NPV of 97.5%, and PPV of 77.0%. When the anti-S titer was greater than 18.1 RU/mL (57.9 BAU/mL), nAbs were positive, with a sensitivity of 90.0% and specificity of 91%. CONCLUSIONS: A titer of SARS-CoV-2 anti-S IgG can be correlated with levels of nAbs.

General bounds on the area under the receiver operating characteristic curve and other performance measures when only a single sensitivity and specificity point is known

Receiver operating characteristic (ROC) curves are often used to quantify the performance of predictive models used in diagnosis, risk stratification and rating systems. The ROC area under the curve (AUC) summarizes the ROC in a single statistic, which also provides a probabilistic interpretation that is isomorphic to the Mann– Whitney–Wilcoxon test. In many settings, such as those involving diagnostic tests for diseases or antibodies, information about the ROC is not reported;instead the true positive. TP / and true negative. TN / rates are reported for a single threshold value. We demonstrate how to calculate the upper and lower bounds for the ROC AUC, given a single. TP;TN / pair. We use simple geometric arguments only, and we present two examples of real-world applications from medicine and finance, involving Covid-19 diagnosis and credit card fraud detection, respectively. In addition, we introduce formally the notion of “pathological” ROC curves and “well-behaved” ROC curves. In the case of well-behaved ROC curves, the bounds on the AUC may be made tighter. In certain special cases involving pathological ROC curves that result from what we term “George Costanza” classifiers, we may transform predictions to obtain well-behaved ROC curves with higher AUC than the original decision process. Our results also enable the calculation of other quantities of interest, such as Cohen’s d or the Pearson correlation between a diagnostic outcome and an actual outcome. These results facilitate the direct comparison of reported performance when model or diagnostic performance is reported for only a single score threshold. © 2022. Infopro Digital Risk (IP) Limited

Sequential Organ Failure Assessment (SOFA) Score and Mortality Prediction in Patients With Severe Respiratory Distress Secondary to COVID-19.

Background This study looks at the validity of the sequential organ failure assessment score (SOFA) in detecting mortality in patients with Coronavirus disease of 2019 (COVID-19) pneumonia. Also, it is looking to determine the optimal SOFA score that will discriminate between mortality and survival. Methods It is a retrospective chart review of the patients admitted to Henry Ford Hospital from March 2020 to December 2020 with COVID-19 pneumonia who developed severe respiratory distress. We collected the following information; patient demographics (age, sex, body mass index), co-morbidities (history of diabetes mellitus, chronic kidney disease, chronic obstructive pulmonary disease, coronary artery disease, or cancer), SOFA scores (the ratio of arterial oxygen tension (PaO2) to the fraction of inspired oxygen, Glasgow Coma Scale (GCS) score, mean arterial pressure, serum creatinine level, bilirubin level, and platelet count) as well as inpatient mortality. Results There were 320 patients; out of these, 111 were intubated. The receiver operating characteristic (ROC) curve for SOFA at the moment of inclusion in the study had an area under the curve of 0.883. The optimal point for discrimination between mortality and survival is SOFA of 5. A SOFA score of less than two is associated with 100% survival, while a score of more than 11 is associated with 100% mortality. Conclusions SOFA score in COVID-19 patients with severe respiratory distress strongly correlates with the initial SOFA score. It is a valuable tool for predicting mortality in COVID-19 patients.

Impact of Acute Confusional State in Patients With COVID-19 and a Predictive Score.

BACKGROUND: Acute confusional state (ACS) in COVID-19 is shown to be associated with poor clinical outcomes. METHODS: We assessed the impact of ACS - defined as a documented deterioration of mental status from baseline on the alertness and orientation to time, place, and person - on inpatient mortality and the need for intensive care unit (ICU) transfer in inpatient admissions with active COVID-19 infection in a single-center retrospective cohort of inpatient admissions from a designated COVID-19 tertiary care center using an electronic health record system. Furthermore, we developed and validated a neurological history and symptom-based predictive score of developing ACS. RESULTS: Thirty seven out of 245 (15%) patients demonstrated ACS. Nineteen (51%) patients had multifactorial ACS, followed by 11 (30%) patients because of hypoxemia. ACS patients were significantly older (80 [70-85] years vs 50.5 [38-69] years, p < 0.001) and demonstrated more frequent history of dementia (43% vs 9%, p < 0.001) and epilepsy (16% vs 2%, p = 0.001). ACS patients observed significantly higher in-hospital mortality (45.9% vs 1.9%, aOR [adjusted odds ratio]: 15.7, 95% CI = 3.6-68.0, p < 0.001) and need for ICU transfer (64.9% vs 35.1%, aOR: 2.7, 95% CI = 1.2-6.1, p = 0.015). In patients who survived hospitalization, ACS was associated with longer hospital stay (6 [3.5-10.5] days vs 3 [2-7] day, p = 0.012) and numerically longer ICU stay (6 [4-10] days vs 3 [2-6] days, p = 0.078). A score to predict ACS demonstrated 75.68% sensitivity and 81.73% specificity at a cutoff of ≥3. CONCLUSION: A high prevalence of ACS was found in patients with COVID-19 in our study cohort. Patients with ACS demonstrated increased mortality and need for ICU care. An internally validated score to predict ACS demonstrated high sensitivity and specificity in our cohort.

Red cell distribution width (RDW): a prognostic indicator of severe COVID-19.

BACKGROUND: The global mortality rate for coronavirus disease 2019 (COVID-19) is 3.68%, but the mortality rate for critically ill patients is as high as 50%. Therefore, the exploration of prognostic predictors for patients with COVID-19 is vital for prompt clinical intervention. Our study aims to explore the predictive value of hematological parameters in the prognosis of patients with severe COVID-19. METHODS: Ninety-eight patients who were diagnosed with COVID-19 at Jingzhou Central Hospital and Central Hospital of Wuhan, Hubei Province, were included in this study. RESULTS: The median age of the patients was 59 [28-80] years; the median age of patients with a good prognosis was 56 [28-79] years, and the median age of patients with a poor outcome was 67 [35-80] years. The patients in the poor outcome group were older than the patients in the good outcome group (P<0.05). The comparison of hematological parameters showed that lymphocyte count (Lym#), red blood cells (RBCs), hemoglobin (HGB), hematocrit (HCT), mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) were significantly lower in the poor outcome group than in the good outcome group (P<0.05). Further, the red cell volume distribution width-CV (RDW-CV) and red cell volume distribution width-SD (RDW-SD) were significantly higher in the poor outcome group than in the good outcome group (P<0.0001). Receiver operating characteristic (ROC) curves showed RDW-SD, with an area under the ROC curve (AUC) of 0.870 [95% confidence interval (CI) 0.796-0.943], was the most significant single parameter for predicting the prognosis of severe patients. When the cut-off value was 42.15, the sensitivity and specificity of RDW-SD for predicting the prognosis of severe patients were 73.1% and 80.2%, respectively. Reticulocyte (RET) channel results showed the RET level was significantly higher in critical patients than in moderate patients and severe patients (P<0.05), which may be one cause of the elevated RDW in patients with a poor outcome. CONCLUSIONS: In this study, the hematological parameters of COVID-19 patients were statistically analyzed. RDW was found to be a prognostic predictor for patients with severe COVID-19, and the increase in RET may contribute to elevated RDW.

Clinical evaluation of fever-screening thermography: impact of consensus guidelines and facial measurement location.

SIGNIFICANCE: Infrared thermographs (IRTs) have been used for fever screening during infectious disease epidemics, including severe acute respiratory syndrome, Ebola virus disease, and coronavirus disease 2019 (COVID-19). Although IRTs have significant potential for human body temperature measurement, the literature indicates inconsistent diagnostic performance, possibly due to wide variations in implemented methodology. A standardized method for IRT fever screening was recently published, but there is a lack of clinical data demonstrating its impact on IRT performance. AIM: Perform a clinical study to assess the diagnostic effectiveness of standardized IRT-based fever screening and evaluate the effect of facial measurement location. APPROACH: We performed a clinical study of 596 subjects. Temperatures from 17 facial locations were extracted from thermal images and compared with oral thermometry. Statistical analyses included calculation of receiver operating characteristic (ROC) curves and area under the curve (AUC) values for detection of febrile subjects. RESULTS: Pearson correlation coefficients for IRT-based and reference (oral) temperatures were found to vary strongly with measurement location. Approaches based on maximum temperatures in either inner canthi or full-face regions indicated stronger discrimination ability than maximum forehead temperature (AUC values of 0.95 to 0.97 versus 0.86 to 0.87, respectively) and other specific facial locations. These values are markedly better than the vast majority of results found in prior human studies of IRT-based fever screening. CONCLUSION: Our findings provide clinical confirmation of the utility of consensus approaches for fever screening, including the use of inner canthi temperatures, while also indicating that full-face maximum temperatures may provide an effective alternate approach.