Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (21 results)

  • Thesis
  • Peer Reviewed
UCLA Electronic Theses and Dissertations (2022)

Air quality is an increasing concern for human health. Among all pollutants, bio-aerosols, which account for 5−34% of indoor particulate matter, are micro-scale airborne living organisms that originate from plants or animals. Inhaled by a human, they can cause irritation, allergies, and various diseases. Conventional detection of bio-aerosols still relies on a pipeline developed for more than 50 years: an aerosol sample is taken at the inspection site, then transmitted to a laboratory and inspected manually under a microscope or through culture experiments. There is still an urgent and unmet need for accurate, label-free and automated bio-aerosol sensing to cover a wide range of species, ideally within a field portable, compact, and cost-effective platform. To this end, we have developed a portable device for air quality monitoring, termed c-Air. It samples aerosols using an impactor, takes holograms of the collected particles using lens-free microscopy and classifies those unlabeled samples using deep learning tools. This device opens new opportunities for environmental sensing and personalized air quality monitoring. This dissertation aims to mature the c-Air device and extend its utility in different real-world applications. A deep learning-assisted workflow for label-free bio-aerosol detection was streamlined, including neural network-based holographic reconstruction and particle classification. The reconstruction network achieves auto focusing and phase recovery simultaneously, greatly reducing the computational complexity. Enabled by this high-efficiency phase recovery method, high throughput monitoring of phase-only, transparent particles becomes feasible. We upgraded the hardware design to monitor the dynamic change of volatile liquid samples such as electronic cigarette (e-cig) generated aerosols using time-lapsed images. Both in-lab and field experiments demonstrated a direct and high-throughput volatility quantification method of e-cig-generated aerosols. And this method can be broadly applied to rapidly characterize various volatile particulate matter. We also developed the next generation of the c-Air device, replacing impactors for particle collection with a virtual impactor to avoid overloading of the sampling cartridge and elongating its lifetime. Combining the virtual impactor with lens-free microscopy, the new device provided volumetric, time-lapsed measurements on flying aerosols. The reconstructed holograms can be further used for particle classification with a neural network. This AI-based bio-aerosol detection and classification device provides a unique solution to indoor air quality monitoring and label-free bio-aerosol sensing.

    Cover page: Label-free Bio-aerosol Detection Using Lens-less Microscopy with Deep Learning
    • Article
    • Peer Reviewed
    UC Irvine Previously Published Works (2016)
    The multi-configurational self-consistent field method is employed to simulate the two-dimensional all-X-ray double-quantum-coherence (XDQC) spectroscopy, a four-wave mixing signal that provides direct signatures of double core hole (DCH) states. The valence electronic structure is probed by capturing the correlation between the single (SCH) and double core hole states. The state-averaged restricted-active-space self-consistent field (SA-RASSCF) approach is used which can treat the valence, SCH, and DCH states at the same theoretical level, and applies to all types of DCHs (located on one or two atoms, K-edge or L-edge), with both accuracy and efficiency. Orbital relaxation introduced by the core hole(s) and the static electron correlation is properly accounted for. The XDQC process can take place via different intermediate DCH state channels by tuning the pulse frequencies. We simulate the XDQC signals for the three isomers of aminophenol at 8 pulse frequency configurations, covering all DCH pathways involving the N1s and O1s core hole (N1sN1s, O1sO1s and N1sO1s), which reveal different patterns of valence excitations.
      Cover page: Study of double core hole excitations in molecules by X-ray double-quantum-coherence signals: a multi-configuration simulationCreative Commons 'BY' version 4.0 license
      • Article
      • Peer Reviewed
      UCLA Previously Published Works (2022)
      The past decade marked a drastic increase in the usage of electronic cigarettes. The adverse health impact of secondhand exposure due to exhaled e-cig particles has raised significant concerns, demanding further research on the characteristics of these particles. In this work, we report direct volatility measurements on exhaled e-cig aerosols using a field-portable device (termed c-Air) enabled by deep learning and lens-free holographic microscopy; for this analysis, we performed a series of field experiments in a vape shop where customers used/vaped their e-cig products. During four days of experiments, we periodically sampled the indoor air with intervals of ~ 16 min and collected the exhaled particles with c-Air. Time-lapse inline holograms of the collected particles were recorded by c-Air and reconstructed using a convolutional neural network yielding phase-recovered microscopic images of the particles. Volumetric decay of individual particles due to evaporation was used as an indicator of the volatility of each aerosol. Volatility dynamics quantified through c-Air experiments showed that indoor vaping increased the percentage of volatile and semi-volatile particles in air. The reported methodology and findings can guide further studies on volatility characterization of indoor e-cig emissions.
        • Article
        • Peer Reviewed
        UCLA Previously Published Works (2022)
        Exposure to bio-aerosols such as pollen can lead to adverse health effects. There is a need for a portable and cost-effective device for long-term monitoring and quantification of various types of pollen. To address this need, we present a mobile and cost-effective label-free sensor that takes holographic images of flowing particulate matter (PM) concentrated by a virtual impactor, which selectively slows down and guides particles larger than 6 μm to fly through an imaging window. The flowing particles are illuminated by a pulsed laser diode, casting their inline holograms on a complementary metal-oxide semiconductor image sensor in a lens-free mobile imaging device. The illumination contains three short pulses with a negligible shift of the flowing particle within one pulse, and triplicate holograms of the same particle are recorded at a single frame before it exits the imaging field-of-view, revealing different perspectives of each particle. The particles within the virtual impactor are localized through a differential detection scheme, and a deep neural network classifies the pollen type in a label-free manner based on the acquired holographic images. We demonstrated the success of this mobile pollen detector with a virtual impactor using different types of pollen (i.e., bermuda, elm, oak, pine, sycamore, and wheat) and achieved a blind classification accuracy of 92.91%. This mobile and cost-effective device weighs ∼700 g and can be used for label-free sensing and quantification of various bio-aerosols over extended periods since it is based on a cartridge-free virtual impactor that does not capture or immobilize PM.
          Cover page: Virtual Impactor-Based Label-Free Pollen Detection using Holography and Deep LearningCreative Commons 'BY-NC-ND' version 4.0 license
          • Article
          • Peer Reviewed
          UCLA Previously Published Works (2021)
          Various volatile aerosols have been associated with adverse health effects; however, characterization of these aerosols is challenging due to their dynamic nature. Here, we present a method that directly measures the volatility of particulate matter (PM) using computational microscopy and deep learning. This method was applied to aerosols generated by electronic cigarettes (e-cigs), which vaporize a liquid mixture (e-liquid) that mainly consists of propylene glycol (PG), vegetable glycerin (VG), nicotine, and flavoring compounds. E-cig-generated aerosols were recorded by a field-portable computational microscope, using an impaction-based air sampler. A lensless digital holographic microscope inside this mobile device continuously records the inline holograms of the collected particles. A deep learning-based algorithm is used to automatically reconstruct the microscopic images of e-cig-generated particles from their holograms and rapidly quantify their volatility. To evaluate the effects of e-liquid composition on aerosol dynamics, we measured the volatility of the particles generated by flavorless, nicotine-free e-liquids with various PG/VG volumetric ratios, revealing a negative correlation between the particles' volatility and the volumetric ratio of VG in the e-liquid. For a given PG/VG composition, the addition of nicotine dominated the evaporation dynamics of the e-cig aerosol and the aforementioned negative correlation was no longer observed. We also revealed that flavoring additives in e-liquids significantly decrease the volatility of e-cig aerosol. The presented holographic volatility measurement technique and the associated mobile device might provide new insights on the volatility of e-cig-generated particles and can be applied to characterize various volatile PM.
            Cover page: Dynamic Imaging and Characterization of Volatile Aerosols in E‑Cigarette Emissions Using Deep Learning-Based Holographic Microscopy
            • Article
            • Peer Reviewed
            UCLA Previously Published Works (2021)
            We demonstrate optical networks composed of diffractive layers trained using deep learning to encode the spatial information of objects into the power spectrum of the diffracted light, which are used to classify objects with a single-pixel spectroscopic detector. Using a plasmonic nanoantenna-based detector, we experimentally validated this single-pixel machine vision framework at terahertz spectrum to optically classify the images of handwritten digits by detecting the spectral power of the diffracted light at ten distinct wavelengths, each representing one class/digit. We also coupled this diffractive network-based spectral encoding with a shallow electronic neural network, which was trained to rapidly reconstruct the images of handwritten digits based on solely the spectral power detected at these ten distinct wavelengths, demonstrating task-specific image decompression. This single-pixel machine vision framework can also be extended to other spectral-domain measurement systems to enable new 3D imaging and sensing modalities integrated with diffractive network-based spectral encoding of information.
              Cover page: Spectrally encoded single-pixel machine vision using diffractive networks
              • Article
              • Peer Reviewed
              UCLA Previously Published Works (2021)
              BACKGROUND: Information and opinions shared by health care providers can affect patient vaccination decisions, but little is known about who health care providers themselves trust for information in the context of new COVID-19 vaccines. OBJECTIVE: The purpose of this study is to investigate which sources of information about COVID-19 vaccines are trusted by health care providers and how they communicate this information to patients. METHODS: This mixed methods study involved a one-time, web-based survey of health care providers and qualitative interviews with a subset of survey respondents. Health care providers (physicians, advanced practice providers, pharmacists, nurses) were recruited from an integrated health system in Southern California using voluntary response sampling, with follow-up interviews with providers who either accepted or declined a COVID-19 vaccine. The outcome was the type of information sources that respondents reported trusting for information about COVID-19 vaccines. Bivariate tests were used to compare trusted information sources by provider type; thematic analysis was used to explore perspectives about vaccine information and communicating with patients about vaccines. RESULTS: The survey was completed by 2948 providers, of whom 91% (n=2683) responded that they had received ≥1 dose of a COVID-19 vaccine. The most frequently trusted source of COVID-19 vaccine information was government agencies (n=2513, 84.2%); the least frequently trusted source was social media (n=691, 9.5%). More physicians trusted government agencies (n=1226, 93%) than nurses (n=927, 78%) or pharmacists (n=203, 78%; P<.001), and more physicians trusted their employer (n=1115, 84%) than advanced practice providers (n=95, 67%) and nurses (n=759, 64%; P=.002). Qualitative themes (n=32 participants) about trusted sources of COVID-19 vaccine information were identified: processing new COVID-19 information in a health care work context likened to a war zone during the pandemic and communicating information to patients. Some providers were hesitant to recommend vaccines to pregnant people and groups they perceived to be at low risk for COVID-19. CONCLUSIONS: Physicians have stronger trust in government sources and their employers for information about COVID-19 vaccines compared with nurses, pharmacists, and advanced practice providers. Strategies such as role modeling, tailored messaging, or talking points with standard language may help providers to communicate accurate COVID-19 vaccine information to patients, and these strategies may also be used with providers with lower levels of trust in reputable information sources.
                Cover page: Health Care Providers Trusted Sources for Information About COVID-19 Vaccines: Mixed Methods Study.
                • Article
                • Peer Reviewed
                UC Irvine Previously Published Works (2019)
                The fluorescence emission from green fluorescent protein (GFP) is known to be heavily influenced by hydrogen bonding between the core fluorophore and the surrounding side chains or water molecules. Yet how to utilize this feature for modulating the fluorescence of GFP chromophore or GFP-like fluorophore still remains elusive. Here we present theoretical calculations to predict how hydrogen bonding could influence the excited states of the GFP-like fluorophores. These studies provide both a new perspective for understanding the photophysical properties of GFP as well as a solid basis for the rational design of GFP-based fluorophores.
                  Cover page: Role of Hydrogen Bonding in Green Fluorescent Protein-like Chromophore EmissionCreative Commons 'BY' version 4.0 license
                  • Article
                  • Peer Reviewed
                  UCLA Previously Published Works (2021)
                  Recent advances in deep learning have been providing non-intuitive solutions to various inverse problems in optics. At the intersection of machine learning and optics, diffractive networks merge wave-optics with deep learning to design task-specific elements to all-optically perform various tasks such as object classification and machine vision. Here, we present a diffractive network, which is used to shape an arbitrary broadband pulse into a desired optical waveform, forming a compact and passive pulse engineering system. We demonstrate the synthesis of various different pulses by designing diffractive layers that collectively engineer the temporal waveform of an input terahertz pulse. Our results demonstrate direct pulse shaping in terahertz spectrum, where the amplitude and phase of the input wavelengths are independently controlled through a passive diffractive device, without the need for an external pump. Furthermore, a physical transfer learning approach is presented to illustrate pulse-width tunability by replacing part of an existing network with newly trained diffractive layers, demonstrating its modularity. This learning-based diffractive pulse engineering framework can find broad applications in e.g., communications, ultra-fast imaging and spectroscopy.
                    Cover page: Terahertz pulse shaping using diffractive surfacesCreative Commons 'BY' version 4.0 license
                    • Article
                    • Peer Reviewed
                    UCLA Previously Published Works (2023)
                    Compartmentalization, leveraging microfluidics, enables highly sensitive assays, but the requirement for significant infrastructure for their design, build, and operation limits access. Multimaterial particle-based technologies thermodynamically stabilize monodisperse droplets as individual reaction compartments with simple liquid handling steps, precluding the need for expensive microfluidic equipment. Here, we further improve the accessibility of this lab on a particle technology to resource-limited settings by combining this assay system with a portable multimodal reader, thus enabling nanoliter droplet assays in an accessible platform. We show the utility of this platform in measuring N-terminal propeptide B-type natriuretic peptide (NT-proBNP), a heart failure biomarker, in complex medium and patient samples. We report a limit of detection of ∼0.05 ng/mL and a linear response between 0.2 and 2 ng/mL in spiked plasma samples. We also show that, owing to the plurality of measurements per sample, swarm sensing acquires better statistical quantitation with a portable reader. Monte Carlo simulations show the increasing capability of this platform to differentiate between negative and positive samples, i.e., below or above the clinical cutoff for acute heart failure (∼0.1 ng/mL), as a function of the number of particles measured. Our platform measurements correlate with gold standard ELISA measurement in cardiac patient samples, and achieve lower variation in measurement across samples compared to the standard well plate-based ELISA. Thus, we show the capabilities of a cost-effective droplet-reader system in accurately measuring biomarkers in nanoliter droplets for diseases that disproportionately affect underserved communities in resource-limited settings.
                      Cover page: Amphiphilic Particle-Stabilized Nanoliter Droplet Reactors with a Multimodal Portable Reader for Distributive Biomarker Quantification.
                      Top