Ultrafast Preliminary Screening of COVID-19 by Machine Learning Analysis of Exhaled NO

Abstract Background A new coronavirus, SARS-CoV-2, has caused the coronavirus disease-2019 (COVID-19) epidemic. Current diagnostic methods mainly include nucleic acid detection, antibody detection, antigen detection, and chest computed tomography (CT) imaging. Although these methods are crucial for the diagnosis of COVID-19, there is a lack of a rapid and economical method for preliminary screening COVID-19.Methods We measured the FeNO concentrations of 103 subjects without COVID-19 and 46 patients with COVID-19. Using machine learning (ML) method, we build a ML model based on fractional exhaled nitric oxide (FeNO) concentration and features of age, and body size for rapid preliminary screening COVID-19 suspects with low-cost.Findings The statistical analysis t-test show that there is a significant difference between the FeNO of healthy people and patients with COVID-19. The ML model can screen out the patients with COVID-19 or other diseases, which show abnormal FeNO distributions. An area under the curve of 0.982 and a sensitivity 0.917 have been achieved for preliminary screening COVID-19 suspects. This non-invasive detection method which takes in two minutes and costs less than a dollar could provide a direction for the control of the rapid spread COVID-19.Interpretation During the COVID-19 pandemic, large numbers and extensive testing of COVID-19 patients remains a problem. Public healthy efforts to limit SARS-CoV-2 spread need to find a more economical and faster screening method.

Download Full-text

Ultrafast Screening of COVID-19 by Machine Learning Analysis of Exhaled NO

10.21203/rs.3.rs-56416/v1 ◽

2020 ◽

Author(s):

Li Yang ◽

Wei Zhou ◽

Jing Ma ◽

Lei Sun ◽

Dong Zhai ◽

...

Keyword(s):

Machine Learning ◽

Healthy Subjects ◽

Detection Efficiency ◽

Low Cost ◽

Diagnostic Methods ◽

Antibody Detection ◽

Nucleic Acid Detection ◽

Disease Costs ◽

Rapid Spread ◽

Multiple Orders

Abstract A new coronavirus, SARS-CoV-2, has caused the coronavirus disease-2019 (COVID-19) epidemic. Current diagnostic methods, including nucleic acid detection, antibody detection, antigen detection and chest computed tomography (CT) imaging, usually take hours, and identification of the disease costs hundreds of dollars. Therefore, an ultrafast and economical detection method is urgently required to control the epidemic spread. Here, we report a rapid and low-cost method for rapidly preliminary screening COVID-19 suspects from healthy people. We established a machine learning (ML) model based on the fractional exhaled nitric oxide (FeNO) concentration, age, sex and body size of 34 COVID-19 patients and 70 healthy subjects. Then, the model was applied to 45 independent subjects, including 12 mild and asymptomatic COVID-19 patients, 10 patients with other diseases, and 23 healthy subjects. The patients with diseases affecting the FeNO including COVID-19, asthma, hypertension and etc were screened out as suspects with the rate of 94.1%. Only one healthy subject was misclassified. This noninvasive and comfortable detection procedure takes in two minutes and costs less than a dollar, which simultaneously improves the detection efficiency and reduces expenses by multiple orders of magnitude. This work may provide a direction for the control of the rapid spread of COVID-19.

Download Full-text

Prevalence and Characterization of Heterogeneous Vancomycin-Intermediate Staphylococcus aureus Isolates from 14 Cities in China

Antimicrobial Agents and Chemotherapy ◽

10.1128/aac.00206-09 ◽

2009 ◽

Vol 53 (9) ◽

pp. 3642-3649 ◽

Cited By ~ 43

Author(s):

Wenjia Sun ◽

Hongbin Chen ◽

Yudong Liu ◽

Chunjiang Zhao ◽

Wright W. Nichols ◽

...

Keyword(s):

Staphylococcus Aureus ◽

Population Analysis ◽

Screening Method ◽

Estimation Method ◽

Area Under The Curve ◽

Brain Heart Infusion ◽

Screening Tests ◽

Screening Methods ◽

Large Numbers ◽

Screening Cascade

ABSTRACT The prevalence of heterogeneous vancomycin-intermediate Staphylococcus aureus (hVISA) among 1,012 vancomycin-susceptible methicillin (meticillin)-resistant S. aureus isolates collected from 14 cities in China from 2005 to 2007 was 13 to 16%, as determined by a combination of (i) measurement by the modified population analysis profile-area under the curve method (PAP-AUC) and (ii) estimation from the measured sensitivity and specificity of a screening method. Two hundred isolates from blood were chosen as a subset for measurement of the sensitivities and the specificities of several previously described screening methods by using the results of PAP-AUC as the reference. During this testing, one isolate was found to be a vancomycin-intermediate S. aureus (VISA) strain so was not used in the evaluation of the screening tests. Of the other 199 isolates, 26 (13.1%) were hVISA, as assessed by PAP-AUC. A screening cascade of culturing the isolates on brain heart infusion agar containing teicoplanin (5 mg/liter) and then subjecting the positive isolates to a macro-Etest method was applied to the 812 non-blood isolates, yielding 149 positive results. From these results and by adjusting for sensitivity (0.423) and specificity (0.861), the prevalence was estimated to be 15.7%. The precision of that estimate was assessed by reapplying the screening cascade to 120 randomly selected isolates from the 812 non-blood isolates and simultaneously determining their heterogeneous vancomycin-intermediate susceptibility status by PAP-AUC. Because PAP-AUC is impractical for use with large numbers of isolates, the screening-based estimation method is useful as a first approximation of the prevalence of hVISA. Of the 27 VISA or hVISA isolates from blood, 22.2% and 74.1% were staphylococcal chromosome cassette mec types II and III, respectively, while 77.8% and 22.2% were agr type 1 and agr type 2, respectively; the MIC ranges were 0.5 to 4 mg/liter for vancomycin and 0.25 to 1 mg/liter for daptomycin.

Download Full-text

Machine learning improves the prediction of febrile neutropenia in Korean inpatients undergoing chemotherapy for breast cancer

Scientific Reports ◽

10.1038/s41598-020-71927-6 ◽

2020 ◽

Vol 10 (1) ◽

Cited By ~ 1

Author(s):

Bum-Joo Cho ◽

Kyoung Min Kim ◽

Sanchir-Erdene Bilegsaikhan ◽

Yong Joon Suh

Keyword(s):

Breast Cancer ◽

Machine Learning ◽

Risk Factors ◽

Febrile Neutropenia ◽

Prediction Models ◽

Learning Algorithms ◽

Area Under The Curve ◽

Primary Prophylaxis ◽

Machine Learning Algorithms ◽

Significant Difference

Abstract Febrile neutropenia (FN) is one of the most concerning complications of chemotherapy, and its prediction remains difficult. This study aimed to reveal the risk factors for and build the prediction models of FN using machine learning algorithms. Medical records of hospitalized patients who underwent chemotherapy after surgery for breast cancer between May 2002 and September 2018 were selectively reviewed for development of models. Demographic, clinical, pathological, and therapeutic data were analyzed to identify risk factors for FN. Using machine learning algorithms, prediction models were developed and evaluated for performance. Of 933 selected inpatients with a mean age of 51.8 ± 10.7 years, FN developed in 409 (43.8%) patients. There was a significant difference in FN incidence according to age, staging, taxane-based regimen, and blood count 5 days after chemotherapy. The area under the curve (AUC) built based on these findings was 0.870 on the basis of logistic regression. The AUC improved by machine learning was 0.908. Machine learning improves the prediction of FN in patients undergoing chemotherapy for breast cancer compared to the conventional statistical model. In these high-risk patients, primary prophylaxis with granulocyte colony-stimulating factor could be considered.

Download Full-text

Evaluation and calibration of a low-cost particle sensor in ambient conditions using machine-learning methods

Atmospheric Measurement Techniques ◽

10.5194/amt-13-1693-2020 ◽

2020 ◽

Vol 13 (4) ◽

pp. 1693-1707 ◽

Cited By ~ 8

Author(s):

Minxing Si ◽

Ying Xiong ◽

Shan Du ◽

Ke Du

Keyword(s):

Machine Learning ◽

Linear Regression ◽

Random Search ◽

Low Cost ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Ambient Conditions ◽

Test Dataset ◽

Compact Size ◽

Significant Difference

Abstract. Particle sensing technology has shown great potential for monitoring particulate matter (PM) with very few temporal and spatial restrictions because of its low cost, compact size, and easy operation. However, the performance of low-cost sensors for PM monitoring in ambient conditions has not been thoroughly evaluated. Monitoring results by low-cost sensors are often questionable. In this study, a low-cost fine particle monitor (Plantower PMS 5003) was colocated with a reference instrument, the Synchronized Hybrid Ambient Real-time Particulate (SHARP) monitor, at the Calgary Varsity air monitoring station from December 2018 to April 2019. The study evaluated the performance of this low-cost PM sensor in ambient conditions and calibrated its readings using simple linear regression (SLR), multiple linear regression (MLR), and two more powerful machine-learning algorithms using random search techniques for the best model architectures. The two machine-learning algorithms are XGBoost and a feedforward neural network (NN). Field evaluation showed that the Pearson correlation (r) between the low-cost sensor and the SHARP instrument was 0.78. The Fligner and Killeen (F–K) test indicated a statistically significant difference between the variances of the PM2.5 values by the low-cost sensor and the SHARP instrument. Large overestimations by the low-cost sensor before calibration were observed in the field and were believed to be caused by the variation of ambient relative humidity. The root mean square error (RMSE) was 9.93 when comparing the low-cost sensor with the SHARP instrument. The calibration by the feedforward NN had the smallest RMSE of 3.91 in the test dataset compared to the calibrations by SLR (4.91), MLR (4.65), and XGBoost (4.19). After calibrations, the F–K test using the test dataset showed that the variances of the PM2.5 values by the NN, XGBoost, and the reference method were not statistically significantly different. From this study, we conclude that a feedforward NN is a promising method to address the poor performance of low-cost sensors for PM2.5 monitoring. In addition, the random search method for hyperparameters was demonstrated to be an efficient approach for selecting the best model structure.

Download Full-text

Laboratory Diagnosis of Human Rabies: Recent Advances

The Scientific World JOURNAL ◽

10.1155/2013/569712 ◽

2013 ◽

Vol 2013 ◽

pp. 1-10 ◽

Cited By ~ 20

Author(s):

Reeta Subramaniam Mani ◽

Shampur Narayan Madhusudana

Keyword(s):

Public Health ◽

Laboratory Diagnosis ◽

Health Impact ◽

Diagnostic Methods ◽

Antibody Detection ◽

Human Rabies ◽

Nucleic Acid Detection ◽

Traditional Methods ◽

Recent Advances ◽

Rabies Diagnosis

Rabies, an acute progressive, fatal encephalomyelitis, transmitted most commonly through the bite of a rabid animal, is responsible for an estimated 61,000 human deaths worldwide. The true disease burden and public health impact due to rabies remain underestimated due to lack of sensitive laboratory diagnostic methods. Rapid diagnosis of rabies can help initiate prompt infection control and public health measures, obviate the need for unnecessary treatment/medical tests, and assist in timely administration of pre- or postexposure prophylactic vaccination to family members and medical staff. Antemortem diagnosis of human rabies provides an impetus for clinicians to attempt experimental therapeutic approaches in some patients, especially after the reported survival of a few cases of human rabies. Traditional methods for antemortem and postmortem rabies diagnosis have several limitations. Recent advances in technology have led to the improvement or development of several diagnostic assays which include methods for rabies viral antigen and antibody detection and assays for viral nucleic acid detection and identification of specific biomarkers. These assays which complement traditional methods have the potential to revolutionize rabies diagnosis in future.

Download Full-text

Digital Biomarker Representing Frailty Phenotypes: The Use of Machine Learning and Sensor-Based Sit-to-Stand Test

Sensors ◽

10.3390/s21093258 ◽

2021 ◽

Vol 21 (9) ◽

pp. 3258

Author(s):

Catherine Park ◽

Ramkinker Mishra ◽

Amir Sharafkhaneh ◽

Mon S. Bryant ◽

Christina Nguyen ◽

...

Keyword(s):

Machine Learning ◽

Low Cost ◽

Wearable Sensors ◽

Area Under The Curve ◽

Frailty Model ◽

Screening Tools ◽

Physical Frailty ◽

Stand Test ◽

Sit To Stand ◽

Range Indicator

Since conventional screening tools for assessing frailty phenotypes are resource intensive and unsuitable for routine application, efforts are underway to simplify and shorten the frailty screening protocol by using sensor-based technologies. This study explores whether machine learning combined with frailty modeling could determine the least sensor-derived features required to identify physical frailty and three key frailty phenotypes (slowness, weakness, and exhaustion). Older participants (n = 102, age = 76.54 ± 7.72 years) were fitted with five wearable sensors and completed a five times sit-to-stand test. Seventeen sensor-derived features were extracted and used for optimal feature selection based on a machine learning technique combined with frailty modeling. Mean of hip angular velocity range (indicator of slowness), mean of vertical power range (indicator of weakness), and coefficient of variation of vertical power range (indicator of exhaustion) were selected as the optimal features. A frailty model with the three optimal features had an area under the curve of 85.20%, a sensitivity of 82.70%, and a specificity of 71.09%. This study suggests that the three sensor-derived features could be used as digital biomarkers of physical frailty and phenotypes of slowness, weakness, and exhaustion. Our findings could facilitate future design of low-cost sensor-based technologies for remote physical frailty assessments via telemedicine.

Download Full-text

Evaluation and Calibration of a Low-cost Particle Sensor in Ambient Conditions Using Machine Learning Technologies

10.5194/amt-2019-393 ◽

2019 ◽

Author(s):

Minxing Si ◽

Ying Xiong ◽

Shan Du ◽

Ke Du

Keyword(s):

Machine Learning ◽

Linear Regression ◽

Random Search ◽

Low Cost ◽

Learning Algorithms ◽

Machine Learning Algorithms ◽

Ambient Conditions ◽

Test Dataset ◽

Compact Size ◽

Significant Difference

Abstract. Particle sensing technology has shown great potential for monitoring particulate matter (PM) with very few temporal and spatial restrictions because of low-cost, compact size, and easy operation. However, the performance of low-cost sensors for PM monitoring in ambient conditions has not been thoroughly evaluated. Monitoring results by low-cost sensors are often questionable. In this study, a low-cost fine particle monitor (Plantower PMS 5003) was co-located with a reference instrument, named Synchronized Hybrid Ambient Real-time Particulate (SHARP) monitor, in Calgary Varsity air monitoring station from December 2018 to April 2019. The study evaluated the performance of this low-cost PM sensor in ambient conditions and calibrated its readings using simple linear regression (SLR), multiple linear regression (MLR), and two more powerful machine learning algorithms using random search techniques for the best model architectures. The two machine learning algorithms are XGBoost and feedforward neural network (NN). Field evaluation showed that the Pearson r between the low-cost sensor and the SHAPR instrument was 0.78. Fligner and Killeen (F-K) test indicated a statistically significant difference between the variances of the PM2.5 values by the low-cost sensor and by the SHARP instrument. Large overestimations by the low-cost sensor before calibration were observed in the field and were believed to be caused by the variation of ambient relative humidity. The root mean square error (RMSE) was 9.93 when comparing the low-cost sensor with the SHARP instrument. The calibration by the feedforward NN had the smallest RMSE of 3.91 in the test dataset, compared to the calibrations by SLR (4.91), MLR (4.65), and XGBoost (4.19). After calibrations, the F-K test using the test dataset showed that the variances of the PM2.5 values by the NN and the XGBoost and by the reference method were not statistically significantly different. From this study, we conclude that feedforward NN is a promising method to address the poor performance of the low-cost sensors for PM2.5 monitoring. In addition, the random search method for hyperparameters was demonstrated to be an efficient approach for selecting the best model structure.

Download Full-text

Anti-Nuclear antibodies: Current concepts and future direction for diagnosing connective tissue disease

Journal of Pathology of Nepal ◽

10.3126/jpn.v5i9.13788 ◽

2015 ◽

Vol 5 (9) ◽

pp. 766-773 ◽

Cited By ~ 2

Author(s):

K Gautam

Keyword(s):

Connective Tissue ◽

False Negative ◽

Screening Method ◽

Connective Tissue Diseases ◽

High Sensitivity ◽

Diagnostic Methods ◽

Antibody Detection ◽

Antinuclear Autoantibodies ◽

Future Direction ◽

High Level

Identification of antinuclear antibodies has been used for the diagnosis of connective tissue diseases for more than fifty years. Indirect immunofluorescence on human epithelial (HEp-2) cells is considered the gold standard screening method for the detection of antinuclear autoantibodies. As the demand of ANA testing increased, the need for automation and standardization has also come forth. A high level of false positive and false negative cases is seen in various populations making it difficult to take clinical decisions. Newer technologies were introduced for the antibody detection to ensure high sensitivity and specificity. This article intends to provide an overview of the concepts on ANA testing, the different diagnostic methods available, the various patterns and clinical utility, the clinical guidelines to be followed, the drawbacks and what lies ahead in the future of ANA testing.Journal of Pathology of Nepal (2015) Vol. 5, 766-773

Download Full-text

Bayesian Network as a Decision Tool for Predicting ALS Disease

Brain Sciences ◽

10.3390/brainsci11020150 ◽

2021 ◽

Vol 11 (2) ◽

pp. 150

Author(s):

Hasan Aykut Karaboga ◽

Aslihan Gunel ◽

Senay Vural Korkut ◽

Ibrahim Demir ◽

Resit Celik

Keyword(s):

Machine Learning ◽

Bayesian Networks ◽

Neurological Diseases ◽

Low Cost ◽

Early Period ◽

Area Under The Curve ◽

Learning Methods ◽

Machine Learning Methods ◽

Comparison Results ◽

Als Patients

Clinical diagnosis of amyotrophic lateral sclerosis (ALS) is difficult in the early period. But blood tests are less time consuming and low cost methods compared to other methods for the diagnosis. The ALS researchers have been used machine learning methods to predict the genetic architecture of disease. In this study we take advantages of Bayesian networks and machine learning methods to predict the ALS patients with blood plasma protein level and independent personal features. According to the comparison results, Bayesian Networks produced best results with accuracy (0.887), area under the curve (AUC) (0.970) and other comparison metrics. We confirmed that sex and age are effective variables on the ALS. In addition, we found that the probability of onset involvement in the ALS patients is very high. Also, a person’s other chronic or neurological diseases are associated with the ALS disease. Finally, we confirmed that the Parkin level may also have an effect on the ALS disease. While this protein is at very low levels in Parkinson’s patients, it is higher in the ALS patients than all control groups.

Download Full-text

GROWTH HORMONE RESPONSES TO FRUCTOSE IN DIABETES MELLITUS

Acta Endocrinologica ◽

10.1530/acta.0.0750497 ◽

1974 ◽

Vol 75 (3) ◽

pp. 497-502

Author(s):

Mayer B. Davidson ◽

Roger M. Steele

Keyword(s):

Diabetes Mellitus ◽

Growth Hormone ◽

Fasting Glucose ◽

Area Under The Curve ◽

Oral Glucose ◽

Gh Secretion ◽

Significant Difference ◽

Duration Of Disease ◽

Hormone Responses

ABSTRACT Since fructose is normally metabolized in diabetics and has recently been shown to stimulate GH secretion, it was used to assess GH responses in diabetics. Fourteen diabetics (9 on insulin) and 8 controls matched for weight were studied. Fructose, infused over 10 min, was compared to arginine, infused over 30 min, both at 0.5 g/kg. Samples were collected at 0, 30, 60, 90 and 120 min and GH responses assessed as area under the curve minus the fasting area. There was no significant difference between the GH responses in diabetics and controls to either agent. Responses to arginine and fructose were significantly correlated (r = 0.60, P < 0.01) in all subjects, but not related to therapy, duration of disease or fasting glucose (75–287 mg/100 ml) in the diabetics. Oral glucose blunted the GH response to fructose in 2 controls. It is concluded that 1) fructose can stimulate GH secretion in male diabetics; 2) however, fructose-stimulated GH responses are not increased in diabetes mellitus.

Download Full-text