Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 6 de 6
Filter
1.
NPJ Digit Med ; 5(1): 5, 2022 Jan 14.
Article in English | MEDLINE | ID: covidwho-1625359

ABSTRACT

While COVID-19 diagnosis and prognosis artificial intelligence models exist, very few can be implemented for practical use given their high risk of bias. We aimed to develop a diagnosis model that addresses notable shortcomings of prior studies, integrating it into a fully automated triage pipeline that examines chest radiographs for the presence, severity, and progression of COVID-19 pneumonia. Scans were collected using the DICOM Image Analysis and Archive, a system that communicates with a hospital's image repository. The authors collected over 6,500 non-public chest X-rays comprising diverse COVID-19 severities, along with radiology reports and RT-PCR data. The authors provisioned one internally held-out and two external test sets to assess model generalizability and compare performance to traditional radiologist interpretation. The pipeline was evaluated on a prospective cohort of 80 radiographs, reporting a 95% diagnostic accuracy. The study mitigates bias in AI model development and demonstrates the value of an end-to-end COVID-19 triage platform.

3.
Clin Infect Dis ; 73(1): e208-e214, 2021 07 01.
Article in English | MEDLINE | ID: covidwho-1291590

ABSTRACT

BACKGROUND: The efficacy of convalescent plasma (CP) for the treatment of coronavirus disease 2019 (COVID-19) remains unclear. METHODS: In a matched cohort analysis of hospitalized patients with severe COVID-19, the impact of CP treatment on in-hospital mortality was evaluated using univariate and multivariate Cox proportional-hazards models, and the impact of CP treatment on time to hospital discharge was assessed using a stratified log-rank analysis. RESULTS: In total, 64 patients who received CP a median of 7 days after symptom onset were compared to a matched control group of 177 patients. The incidence of in-hospital mortality was 12.5% and 15.8% in the CP and control groups, respectively (P = .52). There was no significant difference in the risk of in-hospital mortality between the 2 groups (adjusted hazard ratio [aHR] 0.93, 95% confidence interval [CI] .39-2.20). The overall rate of hospital discharge was not significantly different between the 2 groups (rate ratio [RR] 1.28, 95% CI .91-1.81), although there was a significantly increased rate of hospital discharge among patients 65-years-old or greater who received CP (RR 1.86, 95% CI 1.03-3.36). There was a greater than expected frequency of transfusion reactions in the CP group (2.8% reaction rate observed per unit transfused). CONCLUSIONS: We did not demonstrate a significant difference in risk of mortality or rate of hospital discharge between the CP and control groups. There was a signal for improved outcomes among the elderly, and further adequately powered randomized studies should target this subgroup when assessing the efficacy of CP treatment.


Subject(s)
COVID-19 , Aged , COVID-19/therapy , Cohort Studies , Humans , Immunization, Passive , SARS-CoV-2 , Treatment Outcome
4.
J Med Virol ; 93(2): 916-923, 2021 02.
Article in English | MEDLINE | ID: covidwho-1196420

ABSTRACT

Serology testing of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is increasingly being used during the current pandemic of coronavirus disease 2019 (COVID-19), although its clinical and epidemiologic utilities are still debatable. Characterizing these assays provides scientific basis to best use them. The current study assessed one chemiluminescent assay (Abbott COVID-2 IgG) and two lateral flow assays (STANDARD Q [SQ] IgM/IgG Duo and Wondfo total antibody test) using 113 blood samples from 71 PCR-confirmed COVID-19 hospitalized patients, 119 samples with potential cross-reactions, and 1068 negative controls including 942 pre-pandemic samples. SARS-CoV-2 IgM antibodies became detectable 3-4 days post-symptom onset using SQ IgM test and IgG antibodies were first detected 5-6 days post-onset using SQ IgG. Abbott IgG and Wondfo Total were able to detect antibodies 7 to 8 days post-onset. After 14 days post-symptom onset, the SQ IgG, Abbott IgG and Wondfo Total tests were able to detect antibodies from 100% of the PCR-confirmed patients in this series; 87.5% sensitivity for SQ IgM. Overall agreement was 88.5% between SQ IgM/IgG and Wondfo Total and 94.6% between SQ IgG and Abbott IgG. No cross-reaction due to recent sera with three of the endemic coronaviruses was observed. Viral hepatitis and autoimmune samples were the main source of limited cross-reactions. The specificities were 100% for SQ IgG and Wondfo Total, 99.62% for Abbott IgG, and 98.87% for SQ IgM. These findings demonstrated high sensitivity and specificity of appropriately validated SARS-CoV-2 serologic assays with implications for clinical use and epidemiological seroprevalence studies.


Subject(s)
Antibodies, Viral/blood , COVID-19 Serological Testing/methods , COVID-19/immunology , Aged , COVID-19/diagnosis , Cross Reactions , Female , Humans , Immunoassay/methods , Immunoglobulin G/blood , Immunoglobulin M/blood , Luminescent Measurements/methods , Male , Middle Aged , Reagent Kits, Diagnostic , Sensitivity and Specificity
5.
Lancet Digit Health ; 3(5): e286-e294, 2021 05.
Article in English | MEDLINE | ID: covidwho-1152741

ABSTRACT

BACKGROUND: Chest x-ray is a relatively accessible, inexpensive, fast imaging modality that might be valuable in the prognostication of patients with COVID-19. We aimed to develop and evaluate an artificial intelligence system using chest x-rays and clinical data to predict disease severity and progression in patients with COVID-19. METHODS: We did a retrospective study in multiple hospitals in the University of Pennsylvania Health System in Philadelphia, PA, USA, and Brown University affiliated hospitals in Providence, RI, USA. Patients who presented to a hospital in the University of Pennsylvania Health System via the emergency department, with a diagnosis of COVID-19 confirmed by RT-PCR and with an available chest x-ray from their initial presentation or admission, were retrospectively identified and randomly divided into training, validation, and test sets (7:1:2). Using the chest x-rays as input to an EfficientNet deep neural network and clinical data, models were trained to predict the binary outcome of disease severity (ie, critical or non-critical). The deep-learning features extracted from the model and clinical data were used to build time-to-event models to predict the risk of disease progression. The models were externally tested on patients who presented to an independent multicentre institution, Brown University affiliated hospitals, and compared with severity scores provided by radiologists. FINDINGS: 1834 patients who presented via the University of Pennsylvania Health System between March 9 and July 20, 2020, were identified and assigned to the model training (n=1285), validation (n=183), or testing (n=366) sets. 475 patients who presented via the Brown University affiliated hospitals between March 1 and July 18, 2020, were identified for external testing of the models. When chest x-rays were added to clinical data for severity prediction, area under the receiver operating characteristic curve (ROC-AUC) increased from 0·821 (95% CI 0·796-0·828) to 0·846 (0·815-0·852; p<0·0001) on internal testing and 0·731 (0·712-0·738) to 0·792 (0·780-0 ·803; p<0·0001) on external testing. When deep-learning features were added to clinical data for progression prediction, the concordance index (C-index) increased from 0·769 (0·755-0·786) to 0·805 (0·800-0·820; p<0·0001) on internal testing and 0·707 (0·695-0·729) to 0·752 (0·739-0·764; p<0·0001) on external testing. The image and clinical data combined model had significantly better prognostic performance than combined severity scores and clinical data on internal testing (C-index 0·805 vs 0·781; p=0·0002) and external testing (C-index 0·752 vs 0·715; p<0·0001). INTERPRETATION: In patients with COVID-19, artificial intelligence based on chest x-rays had better prognostic performance than clinical data or radiologist-derived severity scores. Using artificial intelligence, chest x-rays can augment clinical data in predicting the risk of progression to critical illness in patients with COVID-19. FUNDING: Brown University, Amazon Web Services Diagnostic Development Initiative, Radiological Society of North America, National Cancer Institute and National Institute of Biomedical Imaging and Bioengineering of the National Institutes of Health.


Subject(s)
Artificial Intelligence , COVID-19/physiopathology , Prognosis , Radiography, Thoracic , Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Retrospective Studies , SARS-CoV-2 , Severity of Illness Index , Tomography, X-Ray Computed , United States , Young Adult
6.
R I Med J (2013) ; 103(8): 20-23, 2020 Sep 04.
Article in English | MEDLINE | ID: covidwho-755003

ABSTRACT

The rampant COVID-19 pandemic has strained the testing capabilities of healthcare centers across the country. Several nucleic acid and serologic assays are available or currently being developed to meet the growing demand for large-scale testing. This review summarizes the developments of commonly used testing methods and their strategic use in clinical diagnosis and epidemiologic surveillance. This review will cover the basic virology of SARS-CoV-2, nucleic acid amplification testing, serology, antigen testing, as well as newer testing methods such as CRISPR-based assays.


Subject(s)
Betacoronavirus/isolation & purification , Clinical Laboratory Techniques/methods , Coronavirus Infections , Pandemics , Pneumonia, Viral , COVID-19 , COVID-19 Testing , Coronavirus Infections/diagnosis , Coronavirus Infections/epidemiology , Coronavirus Infections/virology , Epidemiological Monitoring , Humans , Pneumonia, Viral/diagnosis , Pneumonia, Viral/epidemiology , Pneumonia, Viral/virology , Procedures and Techniques Utilization , SARS-CoV-2
SELECTION OF CITATIONS
SEARCH DETAIL