skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: A Machine Learning Study of COVID-19 Serology and Molecular Tests and Predictions
Serology and molecular tests are the two most commonly used methods for rapid COVID-19 infection testing. The two types of tests have different mechanisms to detect infection, by measuring the presence of viral SARS-CoV-2 RNA (molecular test) or detecting the presence of antibodies triggered by the SARS-CoV-2 virus (serology test). A handful of studies have shown that symptoms, combined with demographic and/or diagnosis features, can be helpful for the prediction of COVID-19 test outcomes. However, due to nature of the test, serology and molecular tests vary significantly. There is no existing study on the correlation between serology and molecular tests, and what type of symptoms are the key factors indicating the COVID-19 positive tests. In this study, we propose a machine learning based approach to study serology and molecular tests, and use features to predict test outcomes. A total of 2,467 donors, each tested using one or multiple types of COVID-19 tests, are collected as our testbed. By cross checking test types and results, we study correlation between serology and molecular tests. For test outcome prediction, we label 2,467 donors as positive or negative, by using their serology or molecular test results, and create symptom features to represent each donor for learning. Because COVID-19 produces a wide range of symptoms and the data collection process is essentially error prone, we group similar symptoms into bins. This decreases the feature space and sparsity. Using binned symptoms, combined with demographic features, we train five classification algorithms to predict COVID-19 test results. Experiments show that XGBoost achieves the best performance with 76.85% accuracy and 81.4% AUC scores, demonstrating that symptoms are indeed helpful for predicting COVID-19 test outcomes. Our study investigates the relationship between serology and molecular tests, identifies meaningful symptom features associated with COVID-19 infection, and also provides a way for rapid screening and cost effective detection of COVID-19 infection.  more » « less
Award ID(s):
2027339 1763452
PAR ID:
10357379
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Smart health
ISSN:
2352-6483
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; ; ; ; ; et al (, Frontiers in Immunology)
    Infants exposed to caregivers infected with SARS-CoV-2 may have heightened infection risks relative to older children due to their more intensive care and feeding needs. However, there has been limited research on COVID-19 outcomes in exposed infants beyond the neonatal period. Between June 2020 – March 2021, we conducted interviews and collected capillary dried blood spots from 46 SARS-CoV-2 infected mothers and their infants (aged 1-36 months) for up to two months following maternal infection onset (COVID+ group, 87% breastfeeding). Comparative data were also collected from 26 breastfeeding mothers with no known SARS-CoV-2 infection or exposures (breastfeeding control group), and 11 mothers who tested SARS-CoV-2 negative after experiencing symptoms or close contact exposure (COVID- group, 73% breastfeeding). Dried blood spots were assayed for anti-SARS-CoV-2 S-RBD IgG and IgA positivity and anti-SARS-CoV-2 S1 + S2 IgG concentrations. Within the COVID+ group, the mean probability of seropositivity among infant samples was lower than that of corresponding maternal samples (0.54 and 0.87, respectively, for IgG; 0.33 and 0.85, respectively, for IgA), with likelihood of infant infection positively associated with the number of maternal symptoms and other household infections reported. COVID+ mothers reported a lower incidence of COVID-19 symptoms among their infants as compared to themselves and other household adults, and infants had similar PCR positivity rates as other household children. No samples returned by COVID- mothers or their infants tested antibody positive. Among the breastfeeding control group, 44% of mothers but none of their infants tested antibody positive in at least one sample. Results support previous research demonstrating minimal risks to infants following maternal COVID-19 infection, including for breastfeeding infants. 
    more » « less
  2. ; ; ; (, Frontiers in Microbiology)
    COVID-19, caused by SARS-CoV-2, is a systemic illness due to its multiorgan effects in patients. The disease has a detrimental impact on respiratory and cardiovascular systems. One early symptom of infection is anosmia or lack of smell; this implicates the involvement of the olfactory bulb in COVID-19 disease and provides a route into the central nervous system. However, little is known about how SARS-CoV-2 affects neurological or psychological symptoms. SARS-CoV-2 exploits host receptors that converge on pathways that impact psychological symptoms. This systemic review discusses the ways involved by coronavirus infection and their impact on mental health disorders. We begin by briefly introducing the history of coronaviruses, followed by an overview of the essential proteins to viral entry. Then, we discuss the downstream effects of viral entry on host proteins. Finally, we review the literature on host factors that are known to play critical roles in neuropsychiatric symptoms and mental diseases and discuss how COVID-19 could impact mental health globally. Our review details the host factors and pathways involved in the cellular mechanisms, such as systemic inflammation, that play a significant role in the development of neuropsychological symptoms stemming from COVID-19 infection. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ; et al (, Frontiers in Immunology)
    Background Limited data are available regarding the balance of risks and benefits from human milk and/or breastfeeding during and following maternal infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Objective To investigate whether SARS-CoV-2 can be detected in milk and on the breast after maternal coronavirus disease 2019 (COVID-19) diagnosis; and characterize concentrations of milk immunoglobulin (Ig) A specific to the SARS-CoV-2 spike glycoprotein receptor binding domain (RBD) during the 2 months after onset of symptoms or positive diagnostic test. Methods Using a longitudinal study design, we collected milk and breast skin swabs one to seven times from 64 lactating women with COVID-19 over a 2-month period, beginning as early as the week of diagnosis. Milk and breast swabs were analyzed for SARS-CoV-2 RNA, and milk was tested for anti-RBD IgA. Results SARS-CoV-2 was not detected in any milk sample or on 71% of breast swabs. Twenty-seven out of 29 (93%) breast swabs collected after breast washing tested negative for SARS-CoV-2. Detection of SARS-CoV-2 on the breast was associated with maternal coughing and other household COVID-19. Most (75%; 95% CI, 70-79%; n=316) milk samples contained anti-RBD IgA, and concentrations increased ( P =.02) during the first two weeks following onset of COVID-19 symptoms or positive test. Milk-borne anti-RBD IgA persisted for at least two months in 77% of women. Conclusion Milk produced by women with COVID-19 does not contain SARS-CoV-2 and is likely a lasting source of passive immunity via anti-RBD IgA. These results support recommendations encouraging lactating women to continue breastfeeding during and after COVID-19 illness. 
    more » « less
  4. ; ; ; ; ; ; ; ; (, Journal of Breath Research)
    Abstract Rapid testing is essential to fighting pandemics such as coronavirus disease 2019 (COVID-19), the disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Exhaled human breath contains multiple volatile molecules providing powerful potential for non-invasive diagnosis of diverse medical conditions. We investigated breath detection of SARS-CoV-2 infection using cavity-enhanced direct frequency comb spectroscopy (CE-DFCS), a state-of-the-art laser spectroscopic technique capable of a real-time massive collection of broadband molecular absorption features at ro-vibrational quantum state resolution and at parts-per-trillion volume detection sensitivity. Using a total of 170 individual breath samples (83 positive and 87 negative with SARS-CoV-2 based on reverse transcription polymerase chain reaction tests), we report excellent discrimination capability for SARS-CoV-2 infection with an area under the receiver-operating-characteristics curve of 0.849(4). Our results support the development of CE-DFCS as an alternative, rapid, non-invasive test for COVID-19 and highlight its remarkable potential for optical diagnoses of diverse biological conditions and disease states. 
    more » « less
  5. ; ; ; ; ; ; ; ; ; ; et al (, Open Forum Infectious Diseases)
    Abstract BackgroundThe severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron variant has spread globally. However, the contribution of community versus household transmission to the overall risk of infection remains unclear. MethodsBetween November 2021 and March 2022, we conducted an active case-finding study in an urban informal settlement with biweekly visits across 1174 households with 3364 residents. Individuals displaying coronavirus disease 2019 (COVID-19)–related symptoms were identified, interviewed along with household contacts, and defined as index and secondary cases based on reverse-transcription polymerase chain reaction (RT-PCR) and symptom onset. ResultsIn 61 households, we detected a total of 94 RT-PCR–positive cases. Of 69 sequenced samples, 67 cases (97.1%) were attributed to the Omicron BA.1* variant. Among 35 of their households, the secondary attack rate was 50.0% (95% confidence interval [CI], 37.0%–63.0%). Women (relative risk [RR], 1.6 [95% CI, .9–2.7]), older individuals (median difference, 15 [95% CI, 2–21] years), and those reporting symptoms (RR, 1.73 [95% CI, 1.0–3.0]) had a significantly increased risk for SARS-CoV-2 secondary infection. Genomic analysis revealed substantial acquisition of viruses from the community even among households with other SARS-CoV-2 infections. After excluding community acquisition, we estimated a household secondary attack rate of 24.2% (95% CI, 11.9%–40.9%). ConclusionsThese findings underscore the ongoing risk of community acquisition of SARS-CoV-2 among households with current infections. The observed high attack rate necessitates swift booster vaccination, rapid testing availability, and therapeutic options to mitigate the severe outcomes of COVID-19. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email