Machine learning applied to chemical analysis: sensing multiple biomarkers in simulated breath using a temperature-pulsed electronic-nose

Emerging trends in metal oxide-based electronic noses for healthcare applications: a review

An electronic nose (E-nose) is a technology fundamentally inspired by the human nose, designed to detect, recognize, and differentiate specific odors or volatile components in complex and chaotic environments. Comprising an array of sensors with meticulously designed nanostructured architectures, E-noses translate the chemical information captured by these sensors into useful metrics using complex pattern recognition algorithms. E-noses can significantly enhance the quality of life by offering preventive point-of-care devices for medical diagnostics through breath analysis, and by monitoring and tracking hazardous and toxic gases in the environment. They are increasingly being used in defense and surveillance, medical diagnostics, agriculture, environmental monitoring, and product validation and authentication. The major challenge in developing a reliable E-nose involves miniaturization and low power consumption. Various sensing materials are employed to address these issues. This review presents the key advancements over the last decade in E-nose technology, specifically focusing on chemiresistive metal oxide sensing materials. It discusses their sensing mechanisms, integration into portable E-noses, and various data analysis techniques. Additionally, we review the primary metal oxide-based E-noses for disease detection through breath analysis. Finally, we address the major challenges and issues in developing and implementing a portable metal oxide-based E-nose.

Advances in Machine Learning Processing of Big Data from Disease Diagnosis Sensors

Exploring accurate, noninvasive, and inexpensive disease diagnostic sensors is a critical task in the fields of chemistry, biology, and medicine. The complexity of biological systems and the explosive growth of biomarker data have driven machine learning to become a powerful tool for mining and processing big data from disease diagnosis sensors. With the development of bioinformatics and artificial intelligence (AI), machine learning models formed by data mining have been able to guide more sensitive and accurate molecular computing. This review presents an overview of big data collection approaches and fundamental machine learning algorithms and discusses recent advances in machine learning and molecular computational disease diagnostic sensors. More specifically, we highlight existing modular workflows and key opportunities and challenges for machine learning to achieve disease diagnosis through big data mining.

Designing oxide chemiresistors for detecting volatile aromatic compounds: recent progresses and future perspectives

Oxide chemiresistors have mostly been used to detect reactive gases such as ethanol, acetone, formaldehyde, nitric dioxide, and carbon monoxide. However, the selective and sensitive detection of volatile aromatic compounds such as benzene, toluene, and xylene, which are extremely toxic and harmful, using oxide chemiresistors remains challenging because of the molecular stability of benzene rings containing chemicals. Moreover, the performance of the sensing materials is insufficient to detect trace concentration levels of volatile aromatic compounds, which lead to harmful effects on human beings. Here, the strategies for designing highly selective and sensitive volatile aromatic compound gas sensors using oxide chemiresistors were suggested and reviewed. Key approaches include the use of thermal activation, design of sensing materials with high catalytic activity, the utilization of catalytic microreactors and bilayer structures with catalytic overlayer, and the pretreatment of analyte gases or post analysis of sensing signals. In addition, future perspectives from the viewpoint of designing sensing materials and sensor structures for high-performance and robust volatile aromatic compounds gas sensors are provided. Finally, we discuss possible applications of the sensors and sensor arrays.

An Electronic Nose Technology to Quantify Pyrethroid Pesticide Contamination in Tea

The contamination of tea with toxic pesticides is a major concern. Additionally, because of improved detection methods, importers are increasingly rejecting contaminated teas. Here, we describe an electronic nose technique for the rapid detection of pyrethroid pesticides (cyhalothrin, bifenthrin, and fenpropathrin) in tea. Using a PEN 3 electronic nose, the text screened a group of metal oxide sensors and determined that four of them (W5S, W1S, W1W, and W2W) are suitable for the detection of the same pyrethroid pesticide in different concentrations and five of them (W5S, W1S, W1W, W2W, and W2S) are suitable for the detection of pyrethroid pesticide. The models for the determination of cyhalothrin, bifenthrin, and fenpropathrin are established by PLS method. Next, using back propagation (BP) neural network technology, we developed a three-hidden-layer model and a two-hidden-layer model to differentiate among the three pesticides. The accuracy of the three models is 96%, 92%, and 88%, respectively. The recognition accuracies of the three-hidden-layer BP neural network pattern and two-hidden-layer BP neural network pattern are 98.75% and 97.08%, respectively. Our electronic nose system accurately detected and quantified pyrethroid pesticides in tea leaves. We propose that this tool is now ready for practical application in the tea industry.

Nanocomposites SnO₂/SiO₂ for CO Gas Sensors: Microstructure and Reactivity in the Interaction with the Gas Phase

Nanocomposites SnO₂/SiO₂ with a silicon content of [Si]/([Sn] + [Si]) = 3/86 mol.% were obtained by the hydrothermal method. The composition and microstructure of the samples were characterized by EDX, XRD, HRTEM and single-point Brunauer-Emmet-Teller (BET) methods. The surface sites were investigated using thermal analysis, FTIR and XPS. It is shown that the insertion of silicon dioxide up to the value of [Si]/([Sn] + [Si]) = 19 mol.% stabilizes the growth of SnO₂ nanoparticles during high-temperature annealing, which makes it possible to obtain sensor materials operating stably at different temperature conditions. The sensor properties of SnO₂ and SnO₂/SiO₂ nanocomposites were studied by in situ conductivity measurements in the presence of 10–200 ppm CO in dry and humid air in the temperature range of 150–400 °C. It was found that SnO₂/SiO₂ nanocomposites are more sensitive to CO in humid air as compared to pure SnO₂, and the sample with silicon content [Si]/([Sn] + [Si]) = 13 mol.% is resistant to changes in relative air humidity (RH = 4%–65%) in the whole temperature range, which makes it a promising sensor material for detecting CO in real conditions. The results are discussed in terms of the changes in the composition of surface-active groups, which alters the reactivity of the obtained materials.

Multicapillary Gas Chromatography-Temperature Modulated Metal Oxide Semiconductor Sensors Array Detector for Monitoring of Volatile Organic Compounds in Closed Atmosphere Using Gaussian Apodization Factor Analysis

A unique metal oxide semiconductor sensor (MOS) array detector with eight sensors was designed and fabricated in a PTFE chamber as an interface for coupling with multicapillary gas chromatography. This design consists of eight transfer lines with equal length between the multicapillary columns (MCC) and sensors. The deactivated capillary columns were passed through each transfer line and homemade flow splitter to distribute the same gas flow on each sensor. Using the eight ports flow splitter design helps us to equal the length of carrier gas path and flow for each sensor, minimizing the dead volume of the sensor's chamber and increasing chromatographic resolution. In addition to coupling of MCC to MOS array detector and other considerations in hardware design, modulation of MOS temperature was used to increase sensitivity and selectivity, and data analysis was enhanced with adapted Gaussian apodization factor analysis (GAFA) as a multivariate curve resolution algorithm. Continues air sampling and injecting system (CASI) design provides a fast and easily applied method for continues injection of air sample with no additional sample preparation. The analysis cycle time required for each run is less than 300 s. The high sample load and sharp injection with the fast separation by MCC decrease the peak widths and improve detection limits. This homemade customized instrument is an alternative to other time-consuming and expensive technologies for continuous monitoring of outgassing in air samples.

Simultaneous Detection of Two Chemicals Using a TE20-Mode Substrate-Integrated Waveguide Resonator

Microwave resonators working as sensors can detect only a single analyte at a time. To address this issue, a TE₂₀-mode substrate-integrated waveguide (SIW) resonator is exploited, owing to its two distinct regions of high-intensity electric fields, which can be manipulated by loading two chemicals. Two microfluidic channels with unequal fluid-carrying capacities, engraved in a polydimethylsiloxane (PDMS) sheet, can perturb the symmetric electric fields even if loaded with the two extreme cases of dielectric [ethanol (E), deionized water (DI)] and [deionized water, ethanol]. The four layers of the sandwiched structure considered in this study consisted of a top conductive pattern and a bottom ground, both realized on a Rogers RT/Duroid 5880. PDMS-based channels attached with an adhesive serve as the middle layers. The TE₂₀-mode SIW with empty channels resonates at 8.26 GHz and exhibits a -25 dB return loss with an unloaded quality factor of Q ≈ 28. We simultaneously load E and DI and demonstrate the detection of the four possible combinations: [E, DI], [DI, E], [E, E], and [DI, DI]. The performance of our proposed method showed increases in sensitivity (MHz/ε_r) of 7.5%, 216%, and 1170% compared with three previously existing multichannel microwave chemical sensors.

Chemiresistive Electronic Nose toward Detection of Biomarkers in Exhaled Breath

Detection of gas-phase chemicals finds a wide variety of applications, including food and beverages, fragrances, environmental monitoring, chemical and biochemical processing, medical diagnostics, and transportation. One approach for these tasks is to use arrays of highly sensitive and selective sensors as an electronic nose. Here, we present a high performance chemiresistive electronic nose (CEN) based on an array of metal oxide thin films, metal-catalyzed thin films, and nanostructured thin films. The gas sensing properties of the CEN show enhanced sensitive detection of H2S, NH3, and NO in an 80% relative humidity (RH) atmosphere similar to the composition of exhaled breath. The detection limits of the sensor elements we fabricated are in the following ranges: 534 ppt to 2.87 ppb for H2S, 4.45 to 42.29 ppb for NH3, and 206 ppt to 2.06 ppb for NO. The enhanced sensitivity is attributed to the spillover effect by Au nanoparticles and the high porosity of villi-like nanostructures, providing a large surface-to-volume ratio. The remarkable selectivity based on the collection of sensor responses manifests itself in the principal component analysis (PCA). The excellent sensing performance indicates that the CEN can detect the biomarkers of H2S, NH3, and NO in exhaled breath and even distinguish them clearly in the PCA. Our results show high potential of the CEN as an inexpensive and noninvasive diagnostic tool for halitosis, kidney disorder, and asthma.

Investigation of VOCs associated with different characteristics of breast cancer cells

The efficacy of breath volatile organic compounds (VOCs) analysis for the screening of patients bearing breast cancer lesions has been demonstrated by using gas chromatography and artificial olfactory systems. On the other hand, in-vitro studies suggest that VOCs detection could also give important indications regarding molecular and tumorigenic characteristics of tumor cells. Aim of this study was to analyze VOCs in the headspace of breast cancer cell lines in order to ascertain the potentiality of VOCs signatures in giving information about these cells and set-up a new sensor system able to detect breast tumor-associated VOCs. We identified by Gas Chromatography-Mass Spectrometry analysis a VOCs signature that discriminates breast cancer cells for: i) transformed condition; ii) cell doubling time (CDT); iii) Estrogen and Progesterone Receptors (ER, PgR) expression, and HER2 overexpression. Moreover, the signals obtained from a temperature modulated metal oxide semiconductor gas sensor can be classified in order to recognize VOCs signatures associated with breast cancer cells, CDT and ER expression. Our results demonstrate that VOCs analysis could give clinically relevant information about proliferative and molecular features of breast cancer cells and pose the basis for the optimization of a low-cost diagnostic device to be used for tumors characterization.

Application of nanomaterials in microbial-cell biosensor constructions

Microbial cell biosensors, where cells are in direct connection with a transducer enabling quantitative and qualitative detection of an analyte, are very promising analytical tools applied mainly for assays in the environmental field, food industry or biomedicine. Microbial cell biosensors are an excellent alternative to conventional analytical methods due to their specificity, rapid detection and low cost of analysis. Nowadays, nanomaterials are often used in the construction of biosensors to improve their sensitivity and stability. In this review, the combination of microbial and other individual cells with different nanomaterials (carbon nanotubes, graphene, gold nanoparticles, etc.) for the construction of biosensors is described and their applications are provided as well.

Demonstration of fast and accurate discrimination and quantification of chemically similar species utilizing a single cross-selective chemiresistor

Performance characteristics of gas-phase microsensors will determine the ultimate utility of these devices for a wide range of chemical monitoring applications. Commonly employed chemiresistor elements are quite sensitive to selected analytes, and relatively new methods have increased the selectivity to specific compounds, even in the presence of interfering species. Here, we have focused on determining whether purposefully driven temperature modulation can produce faster sensor-response characteristics, which could enable measurements for a broader range of applications involving dynamic compositional analysis. We investigated the response speed of a single chemiresitive In2O3 microhotplate sensor to four analytes (methanol, ethanol, acetone, 2-butanone) by systematically varying the oscillating frequency (semicycle periods of 20-120 ms) of a bilevel temperature cycle applied to the sensing element. It was determined that the fastest response (≈ 9 s), as indicated by a 98% signal-change metric, occurred for a period of 30 ms and that responses under such modulation were dramatically faster than for isothermal operation of the same device (>300 s). Rapid modulation between 150 and 450 °C exerts kinetic control over transient processes, including adsorption, desorption, diffusion, and reaction phenomena, which are important for charge transfer occurring in transduction processes and the observed response times. We also demonstrate that the fastest operation is accompanied by excellent discrimination within a challenging 16-category recognition problem (consisting of the four analytes at four separate concentrations). This critical finding demonstrates that both speed and high discriminatory capabilities can be realized through temperature modulation.

Current breathomics--a review on data pre-processing techniques and machine learning in metabolomics breath analysis

We define breathomics as the metabolomics study of exhaled air. It is a strongly emerging metabolomics research field that mainly focuses on health-related volatile organic compounds (VOCs). Since the amount of these compounds varies with health status, breathomics holds great promise to deliver non-invasive diagnostic tools. Thus, the main aim of breathomics is to find patterns of VOCs related to abnormal (for instance inflammatory) metabolic processes occurring in the human body. Recently, analytical methods for measuring VOCs in exhaled air with high resolution and high throughput have been extensively developed. Yet, the application of machine learning methods for fingerprinting VOC profiles in the breathomics is still in its infancy. Therefore, in this paper, we describe the current state of the art in data pre-processing and multivariate analysis of breathomics data. We start with the detailed pre-processing pipelines for breathomics data obtained from gas-chromatography mass spectrometry and an ion-mobility spectrometer coupled to multi-capillary columns. The outcome of data pre-processing is a matrix containing the relative abundances of a set of VOCs for a group of patients under different conditions (e.g. disease stage, treatment). Independently of the utilized analytical method, the most important question, 'which VOCs are discriminatory?', remains the same. Answers can be given by several modern machine learning techniques (multivariate statistics) and, therefore, are the focus of this paper. We demonstrate the advantages as well the drawbacks of such techniques. We aim to help the community to understand how to profit from a particular method. In parallel, we hope to make the community aware of the existing data fusion methods, as yet unresearched in breathomics.

Assessment, origin, and implementation of breath volatile cancer markers

A new non-invasive and potentially inexpensive frontier in the diagnosis of cancer relies on the detection of volatile organic compounds (VOCs) in exhaled breath samples. Breath can be sampled and analyzed in real-time, leading to fascinating and cost-effective clinical diagnostic procedures. Nevertheless, breath analysis is a very young field of research and faces challenges, mainly because the biochemical mechanisms behind the cancer-related VOCs are largely unknown. In this review, we present a list of 115 validated cancer-related VOCs published in the literature during the past decade, and classify them with respect to their "fat-to-blood" and "blood-to-air" partition coefficients. These partition coefficients provide an estimation of the relative concentrations of VOCs in alveolar breath, in blood and in the fat compartments of the human body. Additionally, we try to clarify controversial issues concerning possible experimental malpractice in the field, and propose ways to translate the basic science results as well as the mechanistic understanding to tools (sensors) that could serve as point-of-care diagnostics of cancer. We end this review with a conclusion and a future perspective.

Detection of volatile organic compounds as biomarkers in breath analysis by different analytical techniques

Breath is a rich mixture containing numerous volatile organic compounds at trace amounts (ppbv–pptv level) such as: hydrocarbons, alcohols, ketones, aldehydes, esters or heterocycles. The presence of some of them depends on health status. Therefore, breath analysis might be useful for clinical diagnostics, therapy monitoring and control of metabolic or biochemical cell cycle products. This Review presents an update on the latest developments in breath analysis applied to diagnosing different diseases with the help of high-quality equipment. Efforts were made to fully and accurately describe traditional and modern techniques used to determine the components of breath. The techniques were compared in terms of design, function and also detection limit of different volatile organic compounds. GC with different detectors, MS, optical sensor and laser spectroscopic detection techniques are also discussed.