Hybrid learning method based on feature clustering and scoring for enhanced COVID-19 breath analysis by an electronic nose

A comprehensive meta-analysis and systematic review of breath analysis in detection of COVID-19 through Volatile organic compounds

BACKGROUND

The COVID-19 pandemic had profound global impacts on daily lives, economic stability, and healthcare systems. Diagnosis of COVID-19 infection via RT-PCR was crucial in reducing spread of disease and informing treatment management. While RT-PCR is a key diagnostic test, there is room for improvement in the development of diagnostic criteria. Identification of volatile organic compounds (VOCs) in exhaled breath provides a fast, reliable, and economically favorable alternative for disease detection.

METHODS

This meta-analysis analyzed the diagnostic performance of VOC-based breath analysis in detection of COVID-19 infection. A systematic review of twenty-nine papers using the grading criteria from Newcastle-Ottawa Scale (NOS) and PRISMA guidelines was conducted.

RESULTS

The cumulative results showed a sensitivity of 0.92 (95 % CI, 90 %-95 %) and a specificity of 0.90 (95 % CI 87 %-93 %). Subgroup analysis by variant demonstrated strong sensitivity to the original strain compared to the Omicron and Delta variant in detection of SARS-CoV-2 infection. An additional subgroup analysis of detection methods showed eNose technology had the highest sensitivity when compared to GC-MS, GC-IMS, and high sensitivity-MS.

CONCLUSION

Overall, these results support the use of breath analysis as a new detection method of COVID-19 infection.

Stacked deep learning approach for efficient SARS-CoV-2 detection in blood samples

Identifying COVID-19 through blood sample analysis is crucial in managing the disease and improving patient outcomes. Despite its advantages, the current test demands certified laboratories, expensive equipment, trained personnel, and 3-4 h for results, with a notable false-negative rate of 15%-20%. This study proposes a stacked deep-learning approach for detecting COVID-19 in blood samples to distinguish uninfected individuals from those infected with the virus. Three stacked deep learning architectures, namely the StackMean, StackMax, and StackRF algorithms, are introduced to improve the detection quality of single deep learning models. To counter the class imbalance phenomenon in the training data, the Synthetic Minority Oversampling Technique (SMOTE) algorithm is also implemented, resulting in increased specificity and sensitivity. The efficacy of the methods is assessed by utilizing blood samples obtained from hospitals in Brazil and Italy. Results revealed that the StackMax method greatly boosted the deep learning and traditional machine learning methods' capability to distinguish COVID-19-positive cases from normal cases, while SMOTE increased the specificity and sensitivity of the stacked models. Hypothesis testing is performed to determine if there is a significant statistical difference in the performance between the compared detection methods. Additionally, the significance of blood sample features in identifying COVID-19 is analyzed using the XGBoost (eXtreme Gradient Boosting) technique for feature importance identification. Overall, this methodology could potentially enhance the timely and precise identification of COVID-19 in blood samples.

Meta-Analysis of COVID-19 Metabolomics Identifies Variations in Robustness of Biomarkers

The global COVID-19 pandemic resulted in widespread harms but also rapid advances in vaccine development, diagnostic testing, and treatment. As the disease moves to endemic status, the need to identify characteristic biomarkers of the disease for diagnostics or therapeutics has lessened, but lessons can still be learned to inform biomarker research in dealing with future pathogens. In this work, we test five sets of research-derived biomarkers against an independent targeted and quantitative Liquid Chromatography-Mass Spectrometry metabolomics dataset to evaluate how robustly these proposed panels would distinguish between COVID-19-positive and negative patients in a hospital setting. We further evaluate a crowdsourced panel comprising the COVID-19 metabolomics biomarkers most commonly mentioned in the literature between 2020 and 2023. The best-performing panel in the independent dataset-measured by F1 score (0.76) and AUROC (0.77)-included nine biomarkers: lactic acid, glutamate, aspartate, phenylalanine, β-alanine, ornithine, arachidonic acid, choline, and hypoxanthine. Panels comprising fewer metabolites performed less well, showing weaker statistical significance in the independent cohort than originally reported in their respective discovery studies. Whilst the studies reviewed here were small and may be subject to confounders, it is desirable that biomarker panels be resilient across cohorts if they are to find use in the clinic, highlighting the importance of assessing the robustness and reproducibility of metabolomics analyses in independent populations.

Enhanced real-time mass spectrometry breath analysis for the diagnosis of COVID-19

Background

Although rapid screening for and diagnosis of coronavirus disease 2019 (COVID-19) are still urgently needed, most current testing methods are long, costly or poorly specific. The objective of the present study was to determine whether or not artificial-intelligence-enhanced real-time mass spectrometry breath analysis is a reliable, safe, rapid means of screening ambulatory patients for COVID-19.

Methods

In two prospective, open, interventional studies in a single university hospital, we used real-time, proton transfer reaction time-of-flight mass spectrometry to perform a metabolomic analysis of exhaled breath from adults requiring screening for COVID-19. Artificial intelligence and machine learning techniques were used to build mathematical models based on breath analysis data either alone or combined with patient metadata.

Results

We obtained breath samples from 173 participants, of whom 67 had proven COVID-19. After using machine learning algorithms to process breath analysis data and further enhancing the model using patient metadata, our method was able to differentiate between COVID-19-positive and -negative participants with a sensitivity of 98%, a specificity of 74%, a negative predictive value of 98%, a positive predictive value of 72% and an area under the receiver operating characteristic curve of 0.961. The predictive performance was similar for asymptomatic, weakly symptomatic and symptomatic participants and was not biased by COVID-19 vaccination status.

Conclusions

Real-time, noninvasive, artificial-intelligence-enhanced mass spectrometry breath analysis might be a reliable, safe, rapid, cost-effective, high-throughput method for COVID-19 screening.

Cluster analysis and forecasting of viruses incidence growth curves: Application to SARS-CoV-2

The sanitary emergency caused by COVID-19 has compromised countries and generated a worldwide health and economic crisis. To provide support to the countries' responses, numerous lines of research have been developed. The spotlight was put on effectively and rapidly diagnosing and predicting the evolution of the pandemic, one of the most challenging problems of the past months. This work contributes to the existing literature by developing a two-step methodology to analyze the transmission rate, designing models applied to territories with similar pandemic behavior characteristics. Virus transmission is considered as bacterial growth curves to understand the spread of the virus and to make predictions about its future evolution. Hence, an analytical clustering procedure is first applied to create groups of locations where the virus transmission rate behaved similarly in the different outbreaks. A curve decomposition process based on an iterative polynomial process is then applied, obtaining meaningful forecasting features. Information of the territories belonging to the same cluster is merged to build models capable of simultaneously predicting the 14-day incidence in several locations using Evolutionary Artificial Neural Networks. The methodology is applied to Andalusia (Spain), although it is applicable to any region across the world. Individual models trained for a specific territory are carried out for comparison purposes. The results demonstrate that this methodology achieves statistically similar, or even better, performance for most of the locations. In addition to being extremely competitive, the main advantage of the proposal lies in its complexity cost reduction. The total number of parameters to be estimated is reduced up to 93.51% for the short term and 93.31% for the mid-term forecasting, respectively. Moreover, the number of required models is reduced by 73.53% and 58.82% for the short- and mid-term forecasting horizons.

Rapid analysis of meat floss origin using a supervised machine learning-based electronic nose towards food authentication

Authentication of meat floss origin has been highly critical for its consumers due to existing potential risks of having allergic diseases or religion perspective related to pork-containing foods. Herein, we developed and assessed a compact portable electronic nose (e-nose) comprising gas sensor array and supervised machine learning with a window time slicing method to sniff and to classify different meat floss products. We evaluated four different supervised learning methods for data classification (i.e., linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), k-nearest neighbors (k-NN), and random forest (RF)). Among them, an LDA model equipped with five-window-extracted feature yielded the highest accuracy values of >99% for both validation and testing data in discriminating beef, chicken, and pork flosses. The obtained e-nose results were correlated and confirmed with the spectral data from Fourier-transform infrared (FTIR) spectroscopy and gas chromatography-mass spectrometry (GC-MS) measurements. We found that beef and chicken had similar compound groups (i.e., hydrocarbons and alcohol). Meanwhile, aldehyde compounds (e.g., dodecanal and 9-octadecanal) were found to be dominant in pork products. Based on its performance evaluation, the developed e-nose system shows promising results in food authenticity testing, which paves the way for ubiquitously detecting deception and food fraud attempts.

Determination of global chemical patterns in exhaled breath for the discrimination of lung damage in postCOVID patients using olfactory technology

The objective of this work was to evaluate the use of an electronic nose and chemometric analysis to discriminate global patterns of volatile organic compounds (VOCs) in breath of postCOVID syndrome patients with pulmonary sequelae. A cross-sectional study was performed in two groups, the group 1 were subjects recovered from COVID-19 without lung damage and the group 2 were subjects recovered from COVID-19 with impaired lung function. The VOCs analysis was executed using a Cyranose 320 electronic nose with 32 sensors, applying principal component analysis (PCA), Partial Least Square-Discriminant Analysis, random forest, canonical discriminant analysis (CAP) and the diagnostic power of the test was evaluated using the ROC (Receiver Operating Characteristic) curve. A total of 228 participants were obtained, for the postCOVID group there are 157 and 71 for the control group, the chemometric analysis results indicate in the PCA an 84% explanation of the variability between the groups, the PLS-DA indicates an observable separation between the groups and 10 sensors related to this separation, by random forest, a classification error was obtained for the control group of 0.090 and for the postCOVID group of 0.088 correct classification. The CAP model showed 83.8% of correct classification and the external validation of the model showed 80.1% of correct classification. Sensitivity and specificity reached 88.9% (73.9%-96.9%) and 96.9% (83.7%-99.9%) respectively. It is considered that this technology can be used to establish the starting point in the evaluation of lung damage in postCOVID patients with pulmonary sequelae.

Potential for Early Noninvasive COVID-19 Detection Using Electronic-Nose Technologies and Disease-Specific VOC Metabolic Biomarkers

The established efficacy of electronic volatile organic compound (VOC) detection technologies as diagnostic tools for noninvasive early detection of COVID-19 and related coronaviruses has been demonstrated from multiple studies using a variety of experimental and commercial electronic devices capable of detecting precise mixtures of VOC emissions in human breath. The activities of numerous global research teams, developing novel electronic-nose (e-nose) devices and diagnostic methods, have generated empirical laboratory and clinical trial test results based on the detection of different types of host VOC-biomarker metabolites from specific chemical classes. COVID-19-specific volatile biomarkers are derived from disease-induced changes in host metabolic pathways by SARS-CoV-2 viral pathogenesis. The unique mechanisms proposed from recent researchers to explain how COVID-19 causes damage to multiple organ systems throughout the body are associated with unique symptom combinations, cytokine storms and physiological cascades that disrupt normal biochemical processes through gene dysregulation to generate disease-specific VOC metabolites targeted for e-nose detection. This paper reviewed recent methods and applications of e-nose and related VOC-detection devices for early, noninvasive diagnosis of SARS-CoV-2 infections. In addition, metabolomic (quantitative) COVID-19 disease-specific chemical biomarkers, consisting of host-derived VOCs identified from exhaled breath of patients, were summarized as possible sources of volatile metabolic biomarkers useful for confirming and supporting e-nose diagnoses.

Influence of dopant concentration on the ammonia sensing performance of citric acid-doped polyvinyl acetate nanofibers

The chemical modification of polymer nanofiber-based ammonia sensors by introducing dopants into the active layers has been proven as one of the low-cost routes to enhance their sensing performance. Herein, we investigate the influence of different citric acid (CA) concentrations on electrospun polyvinyl acetate (PVAc) nanofibers coated on quartz crystal microbalance (QCM) transducers as gravimetric ammonia sensors. The developed CA-doped PVAc nanofiber sensors are tested against various concentrations of ammonia vapors, in which their key sensing performance parameters (i.e., sensitivity, limit of detection (LOD), limit of quantification (LOQ), and repeatability) are studied in detail. The sensitivity and LOD values of 1.34 Hz ppm^-1 and 1 ppm, respectively, can be obtained during ammonia exposure assessment. Adding CA dopants with a higher concentration not only increases the sensor sensitivity linearly, but also prolongs both response and recovery times. This finding allows us to better understand the dopant concentration effect, which subsequently can result in an appropriate strategy for manufacturing high-performance portable nanofiber-based sensing devices.

Fast and noninvasive electronic nose for sniffing out COVID-19 based on exhaled breath-print recognition

The reverse transcription-quantitative polymerase chain reaction (RT-qPCR) approach has been widely used to detect the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). However, instead of using it alone, clinicians often prefer to diagnose the coronavirus disease 2019 (COVID-19) by utilizing a combination of clinical signs and symptoms, laboratory test, imaging measurement (e.g., chest computed tomography scan), and multivariable clinical prediction models, including the electronic nose. Here, we report on the development and use of a low cost, noninvasive method to rapidly sniff out COVID-19 based on a portable electronic nose (GeNose C19) integrating an array of metal oxide semiconductor gas sensors, optimized feature extraction, and machine learning models. This approach was evaluated in profiling tests involving a total of 615 breath samples composed of 333 positive and 282 negative samples. The samples were obtained from 43 positive and 40 negative COVID-19 patients, respectively, and confirmed with RT-qPCR at two hospitals located in the Special Region of Yogyakarta, Indonesia. Four different machine learning algorithms (i.e., linear discriminant analysis, support vector machine, stacked multilayer perceptron, and deep neural network) were utilized to identify the top-performing pattern recognition methods and to obtain a high system detection accuracy (88–95%), sensitivity (86–94%), and specificity (88–95%) levels from the testing datasets. Our results suggest that GeNose C19 can be considered a highly potential breathalyzer for fast COVID-19 screening.