The need for external validation in machine olfaction: emphasis on health-related applications

Real-Time Monitoring of Odour Emissions at the Fenceline of a Waste Treatment Plant by Instrumental Odour Monitoring Systems: Focus on Training Methods

Waste treatment plants (WTPs) often generate odours that may cause nuisance to citizens living nearby. In general, people are becoming more sensitive to environmental issues, and particularly to odour pollution. Instrumental Odour Monitoring Systems (IOMSs) represent an emerging tool for continuous odour measurement and real-time identification of odour peaks, which can provide useful information about the process operation and indicate the occurrence of anomalous conditions likely to cause odour events in the surrounding territories. This paper describes the implementation of two IOMSs at the fenceline of a WTP, focusing on the definition of a specific experimental protocol and data processing procedure for dealing with the interferences of humidity and temperature affecting sensors' responses. Different approaches for data processing were compared and the optimal one was selected based on field performance testing. The humidity compensation model developed proved to be effective, bringing the IOMS classification accuracy above 95%. Also, the adoption of a class-specific regression model compared to a global regression model resulted in an odour quantification capability comparable with those of the reference method (i.e., dynamic olfactometry). Lastly, the validated models were used to process the monitoring data over a period of about one year.

Improved quantitative analysis of tenuifolin using hydrolytic continuous-flow system to build prediction models for its content based on near-infrared spectroscopy

This study used two types of analyses and statistical calculations on powdered samples of Polygala root (PR) and Senega root (SR): (1) determination of saponin content by an independently developed quantitative analysis of tenuifolin content using a flow reactor, and (2) near-infrared spectroscopy (NIR) using crude drug powders as direct samples for metabolic profiling. Furthermore, a prediction model for tenuifolin content was developed and validated using multivariate analysis based on the results of (1) and (2). The goal of this study was to develop a rapid analytical method utilizing the saponin content and explore the possibility of quality control through a wide-area survey of crude drugs using NIR spectroscopy. Consequently, various parameters and appropriate wavelengths were examined in the regression analysis, and a model with a reasonable contribution rate and prediction accuracy was successfully developed. In this case, the wavenumber contributing to the model was consistent with that of tenuifolin, confirming that this model was based on saponin content. In this series of analyses, we have succeeded in developing a model that can quickly estimate saponin content without post-processing and have demonstrated a brief way to perform quality control of crude drugs in the clinical field and on the market.

The Detection of Bacteria in the Maxillary Sinus Secretion of Patients With Acute Rhinosinusitis Using an Electronic Nose: A Pilot Study

OBJECTIVE

Detecting bacteria as a causative pathogen of acute rhinosinusitis (ARS) is a challenging task. Electronic nose technology is a novel method for detecting volatile organic compounds (VOCs) that has also been studied in association with the detection of several diseases. The aim of this pilot study was to analyze maxillary sinus secretion with differential mobility spectrometry (DMS) and to determine whether the secretion demonstrates a different VOC profile when bacteria are present.

METHODS

Adult patients with ARS symptoms were examined. Maxillary sinus contents were aspirated for bacterial culture and DMS analysis. k-Nearest neighbor and linear discriminant analysis were used to classify samples as positive or negative, using bacterial cultures as a reference.

RESULTS

A total of 26 samples from 15 patients were obtained. After leave-one-out cross-validation, k-nearest neighbor produced accuracy of 85%, sensitivity of 67%, specificity of 94%, positive predictive value of 86%, and negative predictive value of 84%.

CONCLUSIONS

The results of this pilot study suggest that bacterial positive and bacterial negative sinus secretion release different VOCs and that DMS has the potential to detect them. However, as the results are based on limited data, further conclusions cannot be made. DMS is a novel method in disease diagnostics and future studies should examine whether the method can detect bacterial ARS by analyzing exhaled air.

External validation of a classifier of daily continuous glucose monitoring (CGM) profiles

As continuous glucose monitoring (CGM) sensors generate ever increasing amounts of CGM data, the need for methods to simplify the storage and analysis of this data becomes increasingly important. Lobo et al. developed a classifier of daily CGM profiles as an initial step in addressing this need. The classifier has several important applications including, but not limited to, data compression, data encryption, and indexing of databases. While the classifier has already successfully classified 99.0% of the 42,595 daily CGM profiles in a Test Set, this work presents an external validation using an external validation set (EVal Set) derived from 8 publicly available data sets. The Test Set and the EVal Set differ in terms of (but not limited to) demographics, data collection time periods, and data collection geographies. The classifier successfully classified 98.2% of the 137,030 daily CGM profiles in the EVal Set. Furthermore, each of the 483 distinct groups of classified daily CGM profiles from the EVal Set retains the same clinical characteristics as the corresponding group from the Test Set, as desired. Finally, the set of unclassified daily CGM profiles from the EVal Set retains the same statistical characteristics as the set of unclassified daily CGM profiles from the Test Set, as desired. These results establish the robustness and generalizability of the classifier: the performance of the classifier is unchanged despite the marked differences between the Test Set and the EVal Set.

Engineering solutions to breath tests based on an e-nose system for silicosis screening and early detection in miners

OBJECTIVES

This study aims to develop an engineering solution to breath tests using an electronic nose (e-nose), and evaluate its diagnosis accuracy for silicosis. Influencing factors of this technique were explored.

METHODS

398 non-silicosis miners and 221 silicosis miners were enrolled in this cross-sectional study. Exhaled breath was analyzed by an array of 16 organic nanofiber sensors along with a customized sample processing system. Principal Component Analysis was used to visualize the breath data, and classifiers were trained by two improved cost-sensitive ensemble algorithms (RF and XGBoost) and two classical algorithms (KNN and SVM). All subjects were included to train the screening model, and an early detection model was run with silicosis cases in stage I. Both 5-fold cross-validation and external validation were adopted. Difference in classifiers caused by algorithms and subjects was quantified using a two-factor analysis of variance. The association between personal smoking habits and classification was investigated by the chi-square test.

RESULTS

Classifiers of ensemble learning performed well in both screening and early detection model, with an accuracy range of 0.817 to 0.987. Classical classifiers showed relatively worse performance. Besides, the ensemble algorithm type and silicosis cases inclusion had no significant effect on classification (p>0.05). There was no connection between personal smoking habits and classification accuracy.

CONCLUSION

Breath tests based on an e-nose consisted of 16x sensor array performed well in silicosis screening and early detection. Raw data input showed a more significant effect on classification compared with the algorithm. Personal smoking habits had little impact on models, supporting the applicability of models in large-scale silicosis screening. The e-nose technique and the breath analysis methods reported are expected to provide a quick and accurate screening for silicosis, and extensible for other diseases.

Breath analysis using electronic nose and gas chromatography-mass spectrometry: A pilot study on bronchial infections in bronchiectasis

BACKGROUND AND AIMS

In this work, breath samples from clinically stable bronchiectasis patients with and without bronchial infections by Pseudomonas Aeruginosa- PA) were collected and chemically analysed to determine if they have clinical value in the monitoring of these patients.

MATERIALS AND METHODS

A cohort was recruited inviting bronchiectasis patients (25) and controls (9). Among the former group, 12 members were suffering PA infection. Breath samples were collected in Tedlar bags and analyzed by e-nose and Gas Chromatography-Mass Spectrometry (GC-MS). The obtained data were analyzed by chemometric methods to determine their discriminant power in regards to their health condition. Results were evaluated with blind samples.

RESULTS

Breath analysis by electronic nose successfully separated the three groups with an overall classification rate of 84% for the three-class classification problem. The best discrimination was obtained between control and bronchiectasis with PA infection samples 100% (CI_95%: 84-100%) on external validation and the results were confirmed by permutation tests. The discrimination analysis by GC-MS provided good results but did not reach proper statistical significance after a permutation test.

CONCLUSIONS

Breath sample analysis by electronic nose followed by proper predictive models successfully differentiated between control, Bronchiectasis and Bronchiectasis PA samples.

Full Workflows for the Analysis of Gas Chromatography-Ion Mobility Spectrometry in Foodomics: Application to the Analysis of Iberian Ham Aroma

Gas chromatography—ion mobility spectrometry (GC-IMS) allows the fast, reliable, and inexpensive chemical composition analysis of volatile mixtures. This sensing technology has been successfully employed in food science to determine food origin, freshness and preventing alimentary fraud. However, GC-IMS data is highly dimensional, complex, and suffers from strong non-linearities, baseline problems, misalignments, peak overlaps, long peak tails, etc., all of which must be corrected to properly extract the relevant features from samples. In this work, a pipeline for signal pre-processing, followed by four different approaches for feature extraction in GC-IMS data, is presented. More precisely, these approaches consist of extracting data features from: (1) the total area of the reactant ion peak chromatogram (RIC); (2) the full RIC response; (3) the unfolded sample matrix; and (4) the ion peak volumes. The resulting pipelines for data processing were applied to a dataset consisting of two different quality class Iberian ham samples, based on their feeding regime. The ability to infer chemical information from samples was tested by comparing the classification results obtained from partial least-squares discriminant analysis (PLS-DA) and the samples’ variable importance for projection (VIP) scores. The choice of a feature extraction strategy is a trade-off between the amount of chemical information that is preserved, and the computational effort required to generate the data models.

Rapid and practical qualitative and quantitative evaluation of non-fumigated ginger and sulfur-fumigated ginger via Fourier-transform infrared spectroscopy and chemometric methods

A strategy was developed to distinguish and quantitate nonfumigated ginger (NS-ginger) and sulfur-fumigated ginger (S-ginger), based on Fourier transform near infrared spectroscopy (FT-NIR) and chemometrics. FT-NIR provided a reliable method to qualitatively assess ginger samples and batches of S-ginger (41) and NS-ginger (39) were discriminated using principal component analysis and orthogonal partial least squares discriminant analysis of FT-NIR data. To generate quantitative methods based on partial least squares (PLS) and counter propagation artificial neural network (CP-ANN) from the FT-NIR, major gingerols were quantified using high performance liquid chromatography (HPLC) and the data used as a reference. Finally, PLS and CP-ANN were deployed to predict concentrations of target compounds in S- and NS-ginger. The results indicated that FT-NIR can provide an alternative to HPLC for prediction of active components in ginger samples and was able to work directly on solid samples.

Reliable chiral recognition with an optoelectronic nose

Chiral discrimination is a key problem in analytical chemistry. It is generally performed using expensive instruments or highly-specific miniaturized sensors. An electronic nose is a bio-inspired instrument capable after training of discriminating a wide variety of analytes. However, generality is achieved at the cost of specificity which makes chiral recognition a challenging task for this kind of device. Recently, a peptide-based optoelectronic nose which can board up to hundreds of different sensing materials has shown promising results, especially in terms of specificity. In line with these results, we describe here its use for chiral recognition. This challenging task requires care, especially in terms of statistical reliability and experimental confounds. For these reasons, we set up an automatic gas sampling system and recorded data over two long sessions, taking care to exclude possible confounds. Two couples of chiral molecules, namely (R) and (S) Limonene and (R) and (S) Carvone, were tested and several statistical analyses indicate the almost perfect discrimination of their two enantiomers. A method to highlight discriminative sensing materials is also proposed and shows that successful discrimination is likely achieved using just a few peptides.

Pattern Recognition and Anomaly Detection by Self-Organizing Maps in a Multi Month E-nose Survey at an Industrial Site

Currently people are aware of the risk related to pollution exposure. Thus odor annoyances are considered a warning about the possible presence of toxic volatile compounds. Malodor often generates immediate alarm among citizens, and electronic noses are convenient instruments to detect mixture of odorant compounds with high monitoring frequency. In this paper we present a study on pattern recognition on ambient air composition in proximity of a gas and oil pretreatment plant by elaboration of data from an electronic nose implementing 10 metal-oxide-semiconductor (MOS) sensors and positioned outdoor continuously during three months. A total of 80,017 e-nose vectors have been elaborated applying the self-organizing map (SOM) algorithm and then k-means clustering on SOM outputs on the whole data set evidencing an anomalous data cluster. Retaining data characterized by dynamic responses of the multisensory system, a SOM with 264 recurrent sensor responses to air mixture sampled at the site and four main air type profiles (clusters) have been identified. One of this sensor profiles has been related to the odor fugitive emissions of the plant, by using ancillary data from a total volatile organic compound (VOC) detector and wind speed and direction data. The overall and daily cluster frequencies have been evaluated, allowing us to identify the daily duration of presence at the monitoring site of air related to industrial emissions. The refined model allowed us to confirm the anomaly detection of the sensor responses.

Diagnosis of ventilator-associated pneumonia using electronic nose sensor array signals: solutions to improve the application of machine learning in respiratory research

Background

Ventilator-associated pneumonia (VAP) is a significant cause of mortality in the intensive care unit. Early diagnosis of VAP is important to provide appropriate treatment and reduce mortality. Developing a noninvasive and highly accurate diagnostic method is important. The invention of electronic sensors has been applied to analyze the volatile organic compounds in breath to detect VAP using a machine learning technique. However, the process of building an algorithm is usually unclear and prevents physicians from applying the artificial intelligence technique in clinical practice. Clear processes of model building and assessing accuracy are warranted. The objective of this study was to develop a breath test for VAP with a standardized protocol for a machine learning technique.

Methods

We conducted a case-control study. This study enrolled subjects in an intensive care unit of a hospital in southern Taiwan from February 2017 to June 2019. We recruited patients with VAP as the case group and ventilated patients without pneumonia as the control group. We collected exhaled breath and analyzed the electric resistance changes of 32 sensor arrays of an electronic nose. We split the data into a set for training algorithms and a set for testing. We applied eight machine learning algorithms to build prediction models, improving model performance and providing an estimated diagnostic accuracy.

Results

A total of 33 cases and 26 controls were used in the final analysis. Using eight machine learning algorithms, the mean accuracy in the testing set was 0.81 ± 0.04, the sensitivity was 0.79 ± 0.08, the specificity was 0.83 ± 0.00, the positive predictive value was 0.85 ± 0.02, the negative predictive value was 0.77 ± 0.06, and the area under the receiver operator characteristic curves was 0.85 ± 0.04. The mean kappa value in the testing set was 0.62 ± 0.08, which suggested good agreement.

Conclusions

There was good accuracy in detecting VAP by sensor array and machine learning techniques. Artificial intelligence has the potential to assist the physician in making a clinical diagnosis. Clear protocols for data processing and the modeling procedure needed to increase generalizability.

Conformal Prediction Based on Raman Spectra for the Classification of Chinese Liquors

This work extends the conventional back-propagation neural network (BPNN) to the classification of Chinese liquors of different flavors according to their Raman spectra. Conformal prediction is applied to assign reliable confidence measures for each classification and support an effective framework to make the machine learning on classification trustable. The BPNN can be used to predict the flavors of Chinese liquors according to their Raman spectra, and a classification rate of 88.96% can be achieved. In order to evaluate each classification, a non-conformity score is defined to generate a P-value for each classification. Moreover, the validity of conformal prediction in online mode is discussed. The number of cumulative errors in the conformal prediction is much less than that without conformal prediction. The relationship between the cumulative error and confidence levels shows that a high confidence level leads to low cumulative errors, but many cumulative errors will occur under a very high confidence level. The result implies that conformal prediction is a useful framework, which can employ classification satisfying a certain level of confidence. Meanwhile, the conformal prediction can improve our classification using a BPNN, when the number of data points is limited.

Electronic Noses in Medical Diagnostics

BACKGROUND

Electronic nose technology is being developed in order to analyse complex mixtures of volatiles in a way parallel to biologic olfaction. When applied in the field of medicine, the use of such devices should enable the identification and discrimination between different diseases. In this review, a comprehensive summary of research in medical diagnostics using electronic noses is presented. A special attention has been paid to the application of these devices and sensor technologies, in response to current trends in medicine.

METHODS

Peer-reviewed research literature pertaining to the subject matter was identified based on a search of bibliographic databases. The quality and relevance of retrieved papers was assessed using standard tools. Their content was critically reviewed and certain information contained therein was compiled in tabularized form.

RESULTS

The majority of reviewed studies show promising results, often surpassing the accuracy and sensitivity of established diagnostic methods. However, only a relatively small number of devices have been field tested. The methods used for sample collection and data processing in various studies were listed in a table, together with electronic nose models used in these investigations.

CONCLUSION

Despite the fact that devices equipped with arrays of chemical sensors are not routinely used in everyday medical practice, their prospective use would solve some established issues in medical diagnostics, as well as lead to developments in prophylactics by facilitating a widespread use of non-invasive screening tests.

Sensors for Lung Cancer Diagnosis

The positive outcome of lung cancer treatment is strongly related to the earliness of the diagnosis. Thus, there is a strong requirement for technologies that could provide an early detection of cancer. The concept of early diagnosis is immediately extended to large population screening, and then, it is strongly related to non-invasiveness and low cost. Sensor technology takes advantage of the microelectronics revolution, and then, it promises to develop devices sufficiently sensitive to detect lung cancer biomarkers. A number of biosensors for the detection of cancer-related proteins have been demonstrated in recent years. At the same time, the interest is growing towards the analysis of volatile metabolites that could be measured directly from the breath. In this paper, a review of the state-of-the-art of biosensors and volatile compound sensors is presented.

A Study of Diagnostic Accuracy Using a Chemical Sensor Array and a Machine Learning Technique to Detect Lung Cancer

Lung cancer is the leading cause of cancer death around the world, and lung cancer screening remains challenging. This study aimed to develop a breath test for the detection of lung cancer using a chemical sensor array and a machine learning technique. We conducted a prospective study to enroll lung cancer cases and non-tumour controls between 2016 and 2018 and analysed alveolar air samples using carbon nanotube sensor arrays. A total of 117 cases and 199 controls were enrolled in the study of which 72 subjects were excluded due to having cancer at another site, benign lung tumours, metastatic lung cancer, carcinoma in situ, minimally invasive adenocarcinoma, received chemotherapy or other diseases. Subjects enrolled in 2016 and 2017 were used for the model derivation and internal validation. The model was externally validated in subjects recruited in 2018. The diagnostic accuracy was assessed using the pathological reports as the reference standard. In the external validation, the areas under the receiver operating characteristic curve (AUCs) were 0.91 (95% CI = 0.79–1.00) by linear discriminant analysis and 0.90 (95% CI = 0.80–0.99) by the supportive vector machine technique. The combination of the sensor array technique and machine learning can detect lung cancer with high accuracy.

In vitro detection of common rhinosinusitis bacteria by the eNose utilising differential mobility spectrometry

Acute rhinosinusitis (ARS) is a sudden, symptomatic inflammation of the nasal and paranasal mucosa. It is usually caused by respiratory virus infection, but bacteria complicate for a small number of ARS patients. The differential diagnostics between viral and bacterial pathogens is difficult and currently no rapid methodology exists, so antibiotics are overprescribed. The electronic nose (eNose) has shown the ability to detect diseases from gas mixtures. Differential mobility spectrometry (DMS) is a next-generation device that can separate ions based on their different mobility in high and low electric fields. Five common rhinosinusitis bacteria (Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, Staphylococcus aureus, and Pseudomonas aeruginosa) were analysed in vitro with DMS. Classification was done using linear discriminant analysis (LDA) and k-nearest neighbour (KNN). The results were validated using leave-one-out cross-validation and separate train and test sets. With the latter, 77% of the bacteria were classified correctly with LDA. The comparative figure with KNN was 79%. In one train-test set, P. aeruginosa was excluded and the four most common ARS bacteria were analysed with LDA and KNN; the correct classification rate was 83 and 85%, respectively. DMS has shown its potential in detecting rhinosinusitis bacteria in vitro. The applicability of DMS needs to be studied with rhinosinusitis patients.

Overoptimism in cross-validation when using partial least squares-discriminant analysis for omics data: a systematic study

Advances in analytical instrumentation have provided the possibility of examining thousands of genes, peptides, or metabolites in parallel. However, the cost and time-consuming data acquisition process causes a generalized lack of samples. From a data analysis perspective, omics data are characterized by high dimensionality and small sample counts. In many scenarios, the analytical aim is to differentiate between two different conditions or classes combining an analytical method plus a tailored qualitative predictive model using available examples collected in a dataset. For this purpose, partial least squares-discriminant analysis (PLS-DA) is frequently employed in omics research. Recently, there has been growing concern about the uncritical use of this method, since it is prone to overfitting and may aggravate problems of false discoveries. In many applications involving a small number of subjects or samples, predictive model performance estimation is only based on cross-validation (CV) results with a strong preference for reporting results using leave one out (LOO). The combination of PLS-DA for high dimensionality data and small sample conditions, together with a weak validation methodology is a recipe for unreliable estimations of model performance. In this work, we present a systematic study about the impact of the dataset size, the dimensionality, and the CV technique used on PLS-DA overoptimism when performance estimation is done in cross-validation. Firstly, by using synthetic data generated from a same probability distribution and with assigned random binary labels, we have obtained a dataset where the true classification rate (CR) is 50%. As expected, our results confirm that internal validation provides overoptimistic estimations of the classification accuracy (i.e., overfitting). We have characterized the CR estimator in terms of bias and variance depending on the internal CV technique used and sample to dimensionality ratio. In small sample conditions, due to the large bias and variance of the estimator, the occurrence of extremely good CRs is common. We have found that overfitting peaks when the sample size in the training subset approaches the feature vector dimensionality minus one. In these conditions, the models are neither under- or overdetermined with a unique solution. This effect is particularly intense for LOO and peaks higher in small sample conditions. Overoptimism is decreased beyond this point where the abundance of noisy produces a regularization effect leading to less complex models. In terms of overfitting, our study ranks CV methods as follows: Bootstrap produces the most accurate estimator of the CR, followed by bootstrapped Latin partitions, random subsampling, K-Fold, and finally, the very popular LOO provides the worst results. Simulation results are further confirmed in real datasets from mass spectrometry and microarrays.

Instrumental drift removal in GC-MS data for breath analysis: the short-term and long-term temporal validation of putative biomarkers for COPD

Breath analysis holds the promise of a non-invasive technique for the diagnosis of diverse respiratory conditions including chronic obstructive pulmonary disease (COPD) and lung cancer. Breath contains small metabolites that may be putative biomarkers of these conditions. However, the discovery of reliable biomarkers is a considerable challenge in the presence of both clinical and instrumental confounding factors. Among the latter, instrumental time drifts are highly relevant, as since question the short and long-term validity of predictive models. In this work we present a methodology to counter instrumental drifts using information from interleaved blanks for a case study of GC-MS data from breath samples. The proposed method includes feature filtering, and additive, multiplicative and multivariate drift corrections, the latter being based on component correction. Biomarker discovery was based on genetic algorithms in a filter configuration using Fisher's ratio computed in the partial least squares-discriminant analysis subspace as a figure of merit. Using our protocol, we have been able to find nine peaks that provide a statistically significant area under the ROC curve of 0.75 for COPD discrimination. The method developed has been successfully validated using blind samples in short-term temporal validation. However, the attempt to use this model for patient screening six months later was not successful. This negative result highlights the importance of increasing validation rigor when reporting biomarker discovery results.

Diagnostic accuracy of breath tests for pneumoconiosis using an electronic nose

Breath analyses have attracted substantial attention as screens for occupational environmental lung disease. The objective of this study was to develop breath tests for pneumoconiosis by analysing volatile organic compounds using an electronic nose. A case-control study was designed. We screened 102 subjects from a cohort of stone workers. After excluding three subjects with poorly controlled diabetes mellitus and one subject with asthma, 98 subjects were enrolled, including 34 subjects with pneumoconiosis and 64 healthy controls. We analysed the subjects' breath using an electronic nose with 32 nanocomposite sensors. Data were randomly split into 80% for model building and 20% for validation. Using a linear discriminate analysis, the sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC) were 67.9%, 88.0%, 80.8%, and 0.91, respectively, in the training set and 66.7%, 71.4%, 70.0%, and 0.86, respectively, in the test set. In subgroup analysis divided by smoking status, the AUROCs for current smokers, former smokers, and subjects who never smoked were 0.94, 0.93, and 0.99, respectively. In subgroup analysis divided by gender, the AUROCs for males and females were 0.95 and 0.99, respectively. Breath tests may have potential as a screen for pneumoconiosis. A multi-centre study is warranted, and the procedures must be standardized before clinical application.

Development of breath test for pneumoconiosis: a case-control study

Background

Lipid peroxidation plays an important role in the pathogenesis of pneumoconiosis. Volatile organic compounds (VOCs) generated from lipid peroxidation might be used to detect pneumoconiosis. The objective of this study was to develop a breath test for pneumoconiosis.

Methods

A case-control study was designed. Breath and ambient air were analysed by gas chromatography/mass spectrometry. After blank correction to prevent contamination from ambient air, we used canonical discriminant analysis (CDA) to assess the discrimination accuracy and principal component analysis (PCA) to generate a prediction score. The prediction accuracy was calculated and validated using the International Classification of Radiographs of the Pneumoconiosis criteria combined with an abnormal pulmonary function test as a reference standard. We generated a receiver operator characteristic (ROC) curve and calculated the area under the ROC curve (AUC) to estimate the screening accuracy of the breath test.

Results

We enrolled 200 stone workers. After excluding 5 subjects with asthma and 16 subjects who took steroids or nonsteroidal anti-inflammatory drugs, a total of 179 subjects were used in the final analyses, which included 25 cases and 154 controls. By CDA, 88.8% of subjects were correctly discriminated by their exposure status and the presence of pneumoconiosis. After excluding the VOCs of automobile exhaust and cigarette smoking, pentane and C5-C7 methylated alkanes constituted the major VOCs in the breath of persons with pneumoconiosis. Using the prediction score generated from PCA, the ROC-AUC was 0.88 (95% CI = 0.80—0.95), and the mean ROC-AUC of 5-fold cross-validation was 0.90. The breath test had good accuracy for pneumoconiosis diagnosis.

Conclusion

The analysis of breath VOCs has potential in the screening of pneumoconiosis for its non-invasiveness and high accuracy. We suggest that a multi-centre study is warranted and that all procedures must be standardized before clinical application.

Conformal Prediction Based on K-Nearest Neighbors for Discrimination of Ginsengs by a Home-Made Electronic Nose

An estimate on the reliability of prediction in the applications of electronic nose is essential, which has not been paid enough attention. An algorithm framework called conformal prediction is introduced in this work for discriminating different kinds of ginsengs with a home-made electronic nose instrument. Nonconformity measure based on k-nearest neighbors (KNN) is implemented separately as underlying algorithm of conformal prediction. In offline mode, the conformal predictor achieves a classification rate of 84.44% based on 1NN and 80.63% based on 3NN, which is better than that of simple KNN. In addition, it provides an estimate of reliability for each prediction. In online mode, the validity of predictions is guaranteed, which means that the error rate of region predictions never exceeds the significance level set by a user. The potential of this framework for detecting borderline examples and outliers in the application of E-nose is also investigated. The result shows that conformal prediction is a promising framework for the application of electronic nose to make predictions with reliability and validity.

Towards localization of malignant sites of asymmetry across bilateral mammograms

BACKGROUND AND OBJECTIVES

The analysis of patterns of asymmetry between the left and right mammograms of a patient can provide meaningful insights into the presence of an underlying tumor in its early stage. However, the identification of breast cancer by investigating bilateral asymmetry is difficult to perform due to the indistinct and borderline nature of the asymmetric signs as they appear on mammograms.

METHODS

In this study, to increase the positive-predictive value of asymmetry in mammographic screening, a novel computerized approach for the automatic localization of malignant sites of asymmetry in mammograms is proposed. The sites of anatomical correspondence between the right and left regions of each radiographic projection were extracted by means of two bilateral masking procedures, inspired by radiologists' criteria in interpreting mammograms and based on the use of detected landmarking structures. Relative variations of spatial patterns of intensity values and of orientations of directional components within each site were quantified by combining multidirectional Gabor filters and indices of structural similarity. The localization of the sites of malignant asymmetry was performed by coupling two quadratic discriminant analysis classifiers, one for each masking procedure, that assigned the likelihood of malignancy to each site of correspondence.

RESULTS

The performance of the proposed method was assessed on 94 mammographic images from two publicly available databases and containing at least one asymmetric site. Sensitivity, specificity and balanced accuracy levels of 0.83 (0.09), 0.75 (0.06), and 0.79 (0.04), respectively were obtained in the classification of malignant asymmetric sites vs benign/normal sites using cross-validation. In addition, a further blind test on a dataset of Full Field Digital Mammograms achieved levels of sensitivity, specificity, and balanced accuracy of 0.86, 0.65, and 0.75, respectively.

CONCLUSIONS

The achieved performance indicates that the proposed system is effective in localizing sites of malignant asymmetry and it is expected to improve computer-aided diagnosis of breast cancer.

Can commercial low-cost sensor platforms contribute to air quality monitoring and exposure estimates?

The emergence of low-cost, user-friendly and very compact air pollution platforms enable observations at high spatial resolution in near-real-time and provide new opportunities to simultaneously enhance existing monitoring systems, as well as engage citizens in active environmental monitoring. This provides a whole new set of capabilities in the assessment of human exposure to air pollution. However, the data generated by these platforms are often of questionable quality. We have conducted an exhaustive evaluation of 24 identical units of a commercial low-cost sensor platform against CEN (European Standardization Organization) reference analyzers, evaluating their measurement capability over time and a range of environmental conditions. Our results show that their performance varies spatially and temporally, as it depends on the atmospheric composition and the meteorological conditions. Our results show that the performance varies from unit to unit, which makes it necessary to examine the data quality of each node before its use. In general, guidance is lacking on how to test such sensor nodes and ensure adequate performance prior to marketing these platforms. We have implemented and tested diverse metrics in order to assess if the sensor can be employed for applications that require high accuracy (i.e., to meet the Data Quality Objectives defined in air quality legislation, epidemiological studies) or lower accuracy (i.e., to represent the pollution level on a coarse scale, for purposes such as awareness raising). Data quality is a pertinent concern, especially in citizen science applications, where citizens are collecting and interpreting the data. In general, while low-cost platforms present low accuracy for regulatory or health purposes they can provide relative and aggregated information about the observed air quality.

Application and Uses of Electronic Noses for Clinical Diagnosis on Urine Samples: A Review

The electronic nose is able to provide useful information through the analysis of the volatile organic compounds in body fluids, such as exhaled breath, urine and blood. This paper focuses on the review of electronic nose studies and applications in the specific field of medical diagnostics based on the analysis of the gaseous headspace of human urine, in order to provide a broad overview of the state of the art and thus enhance future developments in this field. The research in this field is rather recent and still in progress, and there are several aspects that need to be investigated more into depth, not only to develop and improve specific electronic noses for different diseases, but also with the aim to discover and analyse the connections between specific diseases and the body fluids odour. Further research is needed to improve the results obtained up to now; the development of new sensors and data processing methods should lead to greater diagnostic accuracy thus making the electronic nose an effective tool for early detection of different kinds of diseases, ranging from infections to tumours or exposure to toxic agents.

Electronic Nose Testing Procedure for the Definition of Minimum Performance Requirements for Environmental Odor Monitoring

Despite initial enthusiasm towards electronic noses and their possible application in different fields, and quite a lot of promising results, several criticalities emerge from most published research studies, and, as a matter of fact, the diffusion of electronic noses in real-life applications is still very limited. In general, a first step towards large-scale-diffusion of an analysis method, is standardization. The aim of this paper is describing the experimental procedure adopted in order to evaluate electronic nose performances, with the final purpose of establishing minimum performance requirements, which is considered to be a first crucial step towards standardization of the specific case of electronic nose application for environmental odor monitoring at receptors. Based on the experimental results of the performance testing of a commercialized electronic nose type with respect to three criteria (i.e., response invariability to variable atmospheric conditions, instrumental detection limit, and odor classification accuracy), it was possible to hypothesize a logic that could be adopted for the definition of minimum performance requirements, according to the idea that these are technologically achievable.

Sensor Array Design for Complex Sensing Tasks

Chemical detection in complex environments presents numerous challenges for successful implementation. Arrays of sensors are often implemented for complex chemical sensing tasks, but systematic understanding of how individual sensor response characteristics contribute overall detection system performance remains elusive, with generalized strategies for design and optimization of these arrays rarely reported and even less commonly adopted by practitioners. This review focuses on the literature of nonspecific sensor array design and optimization strategies as well as related work that may inform future efforts in complex sensing with arrays.

Breath analysis as a potential and non-invasive frontier in disease diagnosis: an overview

Currently, a small number of diseases, particularly cardiovascular (CVDs), oncologic (ODs), neurodegenerative (NDDs), chronic respiratory diseases, as well as diabetes, form a severe burden to most of the countries worldwide. Hence, there is an urgent need for development of efficient diagnostic tools, particularly those enabling reliable detection of diseases, at their early stages, preferably using non-invasive approaches. Breath analysis is a non-invasive approach relying only on the characterisation of volatile composition of the exhaled breath (EB) that in turn reflects the volatile composition of the bloodstream and airways and therefore the status and condition of the whole organism metabolism. Advanced sampling procedures (solid-phase and needle traps microextraction) coupled with modern analytical technologies (proton transfer reaction mass spectrometry, selected ion flow tube mass spectrometry, ion mobility spectrometry, e-noses, etc.) allow the characterisation of EB composition to an unprecedented level. However, a key challenge in EB analysis is the proper statistical analysis and interpretation of the large and heterogeneous datasets obtained from EB research. There is no standard statistical framework/protocol yet available in literature that can be used for EB data analysis towards discovery of biomarkers for use in a typical clinical setup. Nevertheless, EB analysis has immense potential towards development of biomarkers for the early disease diagnosis of diseases.