Machine Vision Methods, Natural Language Processing, and Machine Learning Algorithms for Automated Dispersion Plot Analysis and Chemical Identification from Complex Mixtures.

Gas-phase trace chemical detection techniques such as ion mobility spectrometry (IMS) and differential mobility spectrometry (DMS) can be used in many settings, such as evaluating the health condition of patients or detecting explosives at airports. These devices separate chemical compounds in a mixture and provide information to identify specific chemical species of interest. Further, these types of devices operate well in both controlled lab environments and in-field applications. Frequently, the commercial versions of these devices are highly tailored for niche applications (e.g., explosives detection) because of the difficulty involved in reconfiguring instrumentation hardware and data analysis software algorithms. In order for researchers to quickly adapt these tools for new purposes and broader panels of chemical targets, it is critical to develop new algorithms and methods for generating libraries of these sensor responses. Microelectromechanical system (MEMS) technology has been used to fabricate DMS devices that miniaturize the platforms for easier deployment; however, concurrent advances in advanced data analytics are lagging. DMS generates complex three-dimensional dispersion plots for both positive and negative ions in a mixture. Although simple spectra of single chemicals are straightforward to interpret (both visually and via algorithms), it is exceedingly challenging to interpret dispersion plots from complex mixtures with many chemical constituents. This study uses image processing and computer vision steps to automatically identify features from DMS dispersion plots. We used the bag-of-words approach adapted from natural language processing and information retrieval to cluster and organize these features. Finally, a support vector machine (SVM) learning algorithm was trained using these features in order to detect and classify specific compounds in these represented conceptualized data outputs. Using this approach, we successfully maintain a high level of correct chemical identification, even when a gas mixture increases in complexity with interfering chemicals present.

High Asymmetric Longitudinal Field Ion Mobility Spectrometry Device for Low Power Mobile Chemical Separation and Detection.

We have developed a novel chemical sensing technique termed high asymmetric longitudinal field ion mobility spectrometry (HALF-IMS), which allows separation of ions based on mobility differences in high and low electric fields. Our device is microfabricated, has a miniature format, and uses exceptionally low power due to the lack of RF separation fields normally associated with ion mobility spectrometry (IMS) or differential mobility spectrometry (DMS). It operates at room temperature and atmospheric pressure. This HALF-IMS chip contains a microscale drift cell where spatially varying electric field regions of high and low strengths are generated by direct current (DC) applied to the electrodes that are physically placed to cause ionic separation as the ionized chemical flows along the drift cell. Power and complexity are reduced at the chip and system levels by reducing the voltage magnitude and using DC-powered electronics. A testing platform utilizing an ultraviolet (UV) photoionization source was used with custom electronic circuit boards to interface with the chip and provide data inputs and outputs. Precise control of the electrode voltages allowed filtering of the passage of the ion of interest through the drift cell, and ionic current was measured at the detector. The device was tested by scanning of electrode voltages and obtaining ion peaks for methyl salicylate, naphthalene, benzene, and 2-butanone. The current experimental setup was capable of detecting as low as ∼80 ppb of methyl salicylate and naphthalene. The use of benzene as a dopant with 2-butanone allowed one to see two ion peaks, corresponding to benzene and 2-butanone.

Portable combination of Fourier transform infrared spectroscopy and differential mobility spectrometry for advanced vapor phase analysis.

Designing mobile devices for the analysis of complex sample mixtures containing a variety of analytes at different concentrations across a large dynamic range remains a challenging task in many analytical scenarios. To meet this challenge, a compact hybrid analytical platform has been developed combining Fourier transform infrared spectroscopy based on substrate-integrated hollow waveguides (iHWG-FTIR) with gas chromatography coupled differential mobility spectrometry (GC-DMS). Due to the complementarity of these techniques regarding analyte type and concentration, their combination provides a promising tool for the detection of complex samples containing a broad range of molecules at different concentrations. To date, the combination of infrared spectroscopy and ion mobility techniques remains expensive and bound to a laboratory utilizing e.g. IMS as prefilter or IR as ionization source. In the present study, a cost-efficient and portable solution has been developed and characterized representing the first truly hyphenated IR-DMS system. As a model analyte mixture, 5 ppm isopropylmercaptan (IPM) in methane (CH4) were diluted, and the concentration-dependent DMS signal of IPM along with the concentration-dependent IR signal of CH4 were recorded for all three hybrid IR-DMS systems. While guiding the sample through the iHWG-FTIR or the GC-DMS first did not affect the obtained signals, optimizing the IR data acquisition parameters did benefit the analytical results.

Detection of Huanglongbing disease using differential mobility spectrometry.

The viability of the multibillion dollar global citrus industry is threatened by the "green menace", citrus greening disease (Huanglongbing, HLB), caused by the bacterial pathogen Candidatus Liberibacter. The long asymptomatic stage of HLB makes it challenging to detect emerging regional infections early to limit disease spread. We have established a novel method of disease detection based on chemical analysis of released volatile organic compounds (VOCs) that emanate from infected trees. We found that the biomarkers "fingerprint" is specific to the causal pathogen and could be interpreted using analytical methods such as gas chromatography/mass spectrometry (GC/MS) and gas chromatography/differential mobility spectrometry (GC/DMS). This VOC-based disease detection method has a high accuracy of ∼90% throughout the year, approaching 100% under optimal testing conditions, even at very early stages of infection where other methods are not adequate. Detecting early infection based on VOCs precedes visual symptoms and DNA-based detection techniques (real-time polymerase chain reaction, RT-PCR) and can be performed at a substantially lower cost and with rapid field deployment.

Automated chemical identification and library building using dispersion plots for differential mobility spectrometry.

Differential mobility spectrometry (DMS) based detectors require rapid data analysis capabilities, embedded into the devices to achieve the optimum detection capabiites as portable trace chemical detectors. Automated algorithm-based DMS dispersion plot data analysis method was applied for the first time to pre-process and separate 3-dimentional (3-D) DMS dispersion data. We previously demonstrated our AnalyzeIMS (AIMS) software was capable of analyzing complex gas chromatography differential mobility spectrometry (GC-DMS) data sets. In our present work, the AIMS software was able to easliy separate DMS dispersion data sets of five chemicals that are important in detection of volatile organic compounds (VOCs): 2-butanone, 2-propanone, ethyl acetate, methanol and ethanol. Identification of chemicals from mixtures, separation of chemicals from a mixture and prediction capability of the software were all tested. These automated algorithms may have potential applications in separation of chemicals (or ion peaks) from other 3-D data obtained by hybrid analytical devices such as mass spectrometry (MS). New algorithm developments are included as future considerations to improve the current numerical approaches to fingerprint chemicals (ions) from a significantly complicated dispersion plot. Comprehensive peak identifcation by DMS-MS, variations of the DMS data due to chemical concentration, gas phase ion chemistry, temperature and pressure of the drift gas are considered in future algorithm improvements.

Modular and reconfigurable gas chromatography / differential mobility spectrometry (GC/DMS) package for detection of volatile organic compounds (VOCs).

Due to the versatility of present day microcontroller boards and open source development environments, new analytical chemistry devices can now be built outside of large industry and instead within smaller individual groups. While there are a wide range of commercial devices available for detecting and identifying volatile organic compounds (VOCs), most of these devices use their own proprietary software and complex custom electronics, making modifications or reconfiguration of the systems challenging. The development of microprocessors for general use, such as the Arduino prototyping platform, now enables custom chemical analysis instrumentation. We have created an example system using commercially available parts, centered around on differential mobility spectrometer (DMS) device. The Modular Reconfigurable Gas Chromatography - Differential Mobility Spectrometry package (MR-GC-DMS) has swappable components allowing it to be quickly reconfigured for specific application purposes as well as broad, generic use. The MR-GC-DMS has a custom user-friendly graphical user interface (GUI) and precisely tuned proportional-integral-derivative controller (PID) feedback control system managing individual temperature-sensitive components. Accurate temperature control programmed into the microcontroller greatly increases repeatability and system performance. Together, this open-source platform enables researchers to quickly combine DMS devices in customized configurations for new chemical sensing applications.

Supervised Semi-Automated Data Analysis Software for Gas Chromatography / Differential Mobility Spectrometry (GC/DMS) Metabolomics Applications.

Modern differential mobility spectrometers (DMS) produce complex and multi-dimensional data streams that allow for near-real-time or post-hoc chemical detection for a variety of applications. An active area of interest for this technology is metabolite monitoring for biological applications, and these data sets regularly have unique technical and data analysis end user requirements. While there are initial publications on how investigators have individually processed and analyzed their DMS metabolomic data, there are no user-ready commercial or open source software packages that are easily used for this purpose. We have created custom software uniquely suited to analyze gas chromatograph / differential mobility spectrometry (GC/DMS) data from biological sources. Here we explain the implementation of the software, describe the user features that are available, and provide an example of how this software functions using a previously-published data set. The software is compatible with many commercial or home-made DMS systems. Because the software is versatile, it can also potentially be used for other similarly structured data sets, such as GC/GC and other IMS modalities.

Coupling a branch enclosure with differential mobility spectrometry to isolate and measure plant volatiles in contained greenhouse settings.

Volatile organic compounds (VOCs) are off-gassed from all living organisms and represent end products of metabolic pathways within the system. In agricultural systems, these VOCs can provide important information on plant health and can ordinarily be measured non-invasively without harvesting tissue from the plants. Previously we reported a portable gas chromatography/differential mobility spectrometry (GC/DMS) system that could distinguish VOC profiles of pathogen-infected citrus from healthy trees before visual symptoms of disease were present. These measurements were taken directly from canopies in the field, but the sampling and analysis protocol did not readily transfer to a controlled greenhouse study where the ambient background air was saturated with volatiles contained in the facility. In this study, we describe for the first time a branch enclosure uniquely coupled with GC/DMS to isolate and measure plant volatiles. To test our system, we sought to replicate our field experiment within a contained greenhouse and distinguish the VOC profiles of healthy versus citrus infected with Candidatus Liberibacter asiaticus. We indeed confirm the ability to track infection-related trace biogenic VOCs using our sampling system and method and we now show this difference in Lisbon lemons (Citrus×limon L. Burm. f.), a varietal not previously reported. Furthermore, the system differentiates the volatile profiles of Lisbon lemons from Washington navels [Citrus sinensis (L.) Osbeck] and also from Tango mandarins (Citrus reticulata Blanco). Based on this evidence, we believe this enclosure-GC/DMS system is adaptable to other volatile-based investigations of plant diseases in greenhouses or other contained settings, and this system may be helpful for basic science research studies of infection mechanisms.

A mobile instrumentation platform to distinguish airway disorders.

Asthma and chronic obstructive pulmonary disease (COPD) are distinct but clinically overlapping airway disorders which often create diagnostic and therapeutic dilemmas. Current strategies to discriminate these diseases are limited by insensitivity and poor performance due to biologic variability. We tested the hypothesis that a gas chromatograph/differential mobility spectrometer (GC/DMS) sensor could distinguish between clinically well-defined groups with airway disorders based on the volatile organic compounds (VOCs) obtained from exhaled breath. After comparing VOC profiles obtained from 13 asthma, 5 COPD and 13 healthy control subjects, we found that VOC profiles distinguished asthma from healthy controls and also a subgroup of asthmatics taking the drug omalizumab from healthy controls. The VOC profiles could not distinguish between COPD and any of the other groups. Our results show a potential application of the GC/DMS for non-invasive and bedside diagnostics of asthma and asthma therapy monitoring. Future studies will focus on larger sample sizes and patient cohorts.