Integrating optical and electrical sensing with machine learning for advanced particle characterization.

Particle classification plays a crucial role in various scientific and technological applications, such as differentiating between bacteria and viruses in healthcare applications or identifying and classifying cancer cells. This technique requires accurate and efficient analysis of particle properties. In this study, we investigated the integration of electrical and optical features through a multimodal approach for particle classification. Machine learning classifier algorithms were applied to evaluate the impact of combining these measurements. Our results demonstrate the superiority of the multimodal approach over analyzing electrical or optical features independently. We achieved an average test accuracy of 94.9% by integrating both modalities, compared to 66.4% for electrical features alone and 90.7% for optical features alone. This highlights the complementary nature of electrical and optical information and its potential for enhancing classification performance. By leveraging electrical sensing and optical imaging techniques, our multimodal approach provides deeper insights into particle properties and offers a more comprehensive understanding of complex biological systems.

Recent Approaches to Design and Analysis of Electrical Impedance Systems for Single Cells Using Machine Learning.

Individual cells have many unique properties that can be quantified to develop a holistic understanding of a population. This can include understanding population characteristics, identifying subpopulations, or elucidating outlier characteristics that may be indicators of disease. Electrical impedance measurements are rapid and label-free for the monitoring of single cells and generate large datasets of many cells at single or multiple frequencies. To increase the accuracy and sensitivity of measurements and define the relationships between impedance and biological features, many electrical measurement systems have incorporated machine learning (ML) paradigms for control and analysis. Considering the difficulty capturing complex relationships using traditional modelling and statistical methods due to population heterogeneity, ML offers an exciting approach to the systemic collection and analysis of electrical properties in a data-driven way. In this work, we discuss incorporation of ML to improve the field of electrical single cell analysis by addressing the design challenges to manipulate single cells and sophisticated analysis of electrical properties that distinguish cellular changes. Looking forward, we emphasize the opportunity to build on integrated systems to address common challenges in data quality and generalizability to save time and resources at every step in electrical measurement of single cells.

Microfluidic Impedance-Deformability Cytometry for Label-Free Single Neutrophil Mechanophenotyping.

The intrinsic biophysical states of neutrophils are associated with immune dysfunctions in diseases. While advanced image-based biophysical flow cytometers can probe cell deformability at high throughput, it is nontrivial to couple different sensing modalities (e.g., electrical) to measure other critical cell attributes including cell viability and membrane integrity. Herein, an "optics-free" impedance-deformability cytometer for multiparametric single cell mechanophenotyping is reported. The microfluidic platform integrates hydrodynamic cell pinching, and multifrequency impedance quantification of cell size, deformability, and membrane impedance (indicative of cell viability and activation). A newly-defined "electrical deformability index" is validated by numerical simulations, and shows strong correlations with the optical cell deformability index of HL-60 experimentally. Human neutrophils treated with various biochemical stimul are further profiled, and distinct differences in multimodal impedance signatures and UMAP analysis are observed. Overall, the integrated cytometer enables label-free cell profiling at throughput of >1000 cells min-1 without any antibodies labeling to facilitate clinical diagnostics.

Enhancing signals of microfluidic impedance cytometry through optimization of microelectrode array.

Microfluidic impedance cytometry shows a great value in biomedical diagnosis. However, the crosstalk between neighboring microelectrodes strongly weakens the impedance signal. Hereby, we demonstrate a novel microfluidic impedance cytometer consisted of sensing electrodes and ground electrodes (GNDs). The simulation reveals a signal enhancement by more than five times with GNDs compared to that without ones. We also found that the linear correlation between the impedance at a high frequency and that at a low frequency varies as microparticle size changes, which can be used for microparticle classification. The study can help with microelectrode optimization and signal processing for microfluidic impedance analysis.

Digital filtering dissemination for optimizing impedance cytometry signal quality and counting accuracy.

Improving biosensor performance which utilize impedance cytometry is a highly interested research topic for many clinical and diagnostic settings. During development, a sensor's design and external factors are rigorously optimized, but improvements in signal quality and interpretation are usually still necessary to produce a sensitive and accurate product. A common solution involves digital signal processing after sample analysis, but these methods frequently fall short in providing meaningful signal outcome changes. This shortcoming may arise from a lack of investigative research into selecting and using signal processing functions, as many choices in current sensors are based on either theoretical results or estimated hypotheses. While a ubiquitous condition set is improbable across diverse impedance cytometry designs, there lies a need for a streamlined and rapid analytical method for discovering those conditions for unique sensors. Herein, we present a comprehensive dissemination of digital filtering parameters applied on experimental impedance cytometry data for determining the limits of signal processing on signal quality improvements. Various filter orders, cutoff frequencies, and filter types are applied after data collection for highest achievable noise reduction. After designing and fabricating a microfluidic impedance cytometer, 9 µm polystyrene particles were measured under flow and signal quality improved by 6.09 dB when implementing digital filtering. This approached was then translated to isolated human neutrophils, where similarly, signal quality improved by 7.50 dB compared to its unfiltered original data. By sweeping all filtering conditions and devising a system to evaluate filtering performance both by signal quality and object counting accuracy, this may serve as a framework for future systems to determine their appropriately optimized filtering configuration.

Identification of Single Yeast Budding Using Impedance Cytometry with a Narrow Electrode Span.

Impedance cytometry is wildly used in single-cell detection, and its sensitivity is essential for determining the status of single cells. In this work, we focus on the effect of electrode gap on detection sensitivity. Through comparing the electrode span of 1 µm and 5 µm, our work shows that narrowing the electrode span could greatly improve detection sensitivity. The mechanism underlying the sensitivity improvement was analyzed via numerical simulation. The small electrode gap (1 µm) allows the electric field to concentrate near the detection area, resulting in a high sensitivity for tiny particles. This finding is also verified with the mixture suspension of 1 µm and 3 µm polystyrene beads. As a result, the electrodes with 1 µm gap can detect more 1 µm beads in the suspension than electrodes with 5 µm gap. Additionally, for single yeast cells analysis, it is found that impedance cytometry with 1 µm electrodes gap can easily distinguish budding yeast cells, which cannot be realized by the impedance cytometry with 5 µm electrodes gap. All experimental results support that narrowing the electrode gap is necessary for tiny particle detection, which is an important step in the development of submicron and nanoscale impedance cytometry.

Direct and Label-Free Cell Status Monitoring of Spheroids and Microcarriers Using Microfluidic Impedance Cytometry.

3D cellular spheroids/microcarriers (100 µm-1 mm) are widely used in biomanufacturing, and non-invasive biosensors are useful to monitor cell quality in bioprocesses. In this work, a novel microfluidic approach for label-free and continuous-flow monitoring of single spheroid/microcarrier (hydrogel and Cytodex) based on electrical impedance spectroscopy using co-planar Field's metal electrodes is reported. Through numerical simulation and experimental validation, two unique impedance signatures (|ZLF | (60 kHz), |ZHF | (1 MHz)) which are optimal for spheroid growth and viability monitoring are identified. Using a closed-loop recirculation system, it is demonstrated that |ZLF | increases with breast cancer (MCF-7) spheroid biomass, while higher opacity (impedance ratio |ZHF |/|ZLF |) indicates cell death due to compromised cell membrane. Anti-cancer drug (paclitaxel)-treated spheroids also exhibit lower |ZLF | with increased cell dissociation. Interestingly, impedance characterization of adipose-derived mesenchymal stem cell differentiation on Cytodex microcarriers reveals that adipogenic cells (higher intracellular lipid content) exhibit higher impedance than osteogenic cells (more conductive due to calcium ions) for both microcarriers and single cell level. Taken together, the developed platform offers great versatility for multi-parametric analysis of spheroids/microcarriers at high throughput (≈1 particle/s), and can be readily integrated into bioreactors for long-term and remote monitoring of biomass and cell quality.

A neural network approach for real-time particle/cell characterization in microfluidic impedance cytometry.

Microfluidic applications such as active particle sorting or selective enrichment require particle classification techniques that are capable of working in real time. In this paper, we explore the use of neural networks for fast label-free particle characterization during microfluidic impedance cytometry. A recurrent neural network is designed to process data from a novel impedance chip layout for enabling real-time multiparametric analysis of the measured impedance data streams. As demonstrated with both synthetic and experimental datasets, the trained network is able to characterize with good accuracy size, velocity, and cross-sectional position of beads, red blood cells, and yeasts, with a unitary prediction time of 0.4 ms. The proposed approach can be extended to other device designs and cell types for electrical parameter extraction. This combination of microfluidic impedance cytometry and machine learning can serve as a stepping stone to real-time single-cell analysis and sorting. Graphical Abstract.

Toward Microfluidic Label-Free Isolation and Enumeration of Circulating Tumor Cells from Blood Samples.

The isolation, analysis, and enumeration of circulating tumor cells (CTCs) from cancer patient blood samples are a paradigm shift for cancer patient diagnosis, prognosis, and treatment monitoring. Most methods used to isolate and enumerate these target cells rely on the expression of cell surface markers, which varies between patients, cancer types, tumors, and stages. Here, we propose a label-free high-throughput platform to isolate, enumerate, and size CTCs on two coupled microfluidic devices. Cancer cells were purified through a Vortex chip and subsequently flowed in-line to an impedance chip, where a pair of electrodes measured fluctuations of an applied electric field generated by cells passing through. A proof-of-concept of the coupling of those two devices was demonstrated with beads and cells. First, the impedance chip was tested as a stand-alone device: (1) with beads (mean counting error of 1.0%, sizing information clearly separated three clusters for 8, 15, and 20 um beads, respectively) as well as (2) with cancer cells (mean counting error of 3.5%). Second, the combined setup was tested with beads, then with cells in phosphate-buffered saline, and finally with cancer cells spiked in healthy blood. Experiments demonstrated that the Vortex HT chip enriched the cancer cells, which then could be counted and differentiated from smaller blood cells by the impedance chip based on size information. Further discrimination was shown with dual high-frequency measurements using electric opacity, highlighting the potential application of this combined setup for a fully integrated label-free isolation and enumeration of CTCs from cancer patient samples. © 2019 International Society for Advancement of Cytometry.

Impedance-based viscoelastic flow cytometry.

Elastic nature of the viscoelastic fluids induces lateral migration of particles into a single streamline and can be used by microfluidic based flow cytometry devices. In this study, we investigated focusing efficiency of polyethylene oxide based viscoelastic solutions at varying ionic concentration to demonstrate their use in impedimetric particle characterization systems. Rheological properties of the viscoelastic fluid and particle focusing performance are not affected by ionic concentration. We investigated the viscoelastic focusing dynamics using polystyrene (PS) beads and human red blood cells (RBCs) suspended in the viscoelastic fluid. Elasto-inertial focusing of PS beads was achieved with the combination of inertial and viscoelastic effects. RBCs were aligned along the channel centerline in parachute shape which yielded consistent impedimetric signals. We compared our impedance-based microfluidic flow cytometry results for RBCs and PS beads by analyzing particle transit time and peak amplitude at varying viscoelastic focusing conditions obtained at different flow rates. We showed that single orientation, single train focusing of nonspherical RBCs can be achieved with polyethylene oxide based viscoelastic solution that has been shown to be a good candidate as a carrier fluid for impedance cytometry.

A portable battery powered microfluidic impedance cytometer with smartphone readout: towards personal health monitoring.

We present a portable system for personalized blood cell counting consisting of a microfluidic impedance cytometer and portable analog readout electronics, feeding into an analog-to-digital converter (ADC), and being transmitted via Bluetooth to a user-accessible mobile application. We fabricated a microfluidic impedance cytometer with a novel portable analog readout. The novel design of the analog readout, which consists of a lock-in-amplifier followed by a high-pass filter stage for subtraction of drift and DC offset, and a post-subtraction high gain stage, enables detection of particles and cells as small as 1 µm in diameter, despite using a low-end 8-bit ADC. The lock-in-amplifier and the ADC were set up to receive and transmit data from a Bluetooth module. In order to initiate the system, as well as to transmit all of the data, a user friendly mobile application was developed, and a proof-of-concept trial was run on a blood sample. Applications such as personalized health monitoring require robust device operation and resilience to clogging. It is desirable to avoid using channels comparable in size to the particles being detected thus requiring high levels of sensitivity. Despite using low-end off-the-shelf hardware, our sensing platform was capable of detecting changes in impedance as small as 0.032%, allowing detection of 3 µm diameter particles in a 300 µm wide channel. The sensitivity of our system is comparable to that of a high-end bench-top impedance spectrometer when tested using the same sensors. The novel analog design allowed for an instrument with a footprint of less than 80 cm2. The aim of this work is to demonstrate the potential of using microfluidic impedance spectroscopy for low cost health monitoring. We demonstrated the utility of the platform technology towards cell counting, however, our platform is broadly applicable to assaying wide panels of biomarkers including proteins, nucleic acids, and various cell types.

PicoMolar level detection of protein biomarkers based on electronic sizing of bead aggregates: theoretical and experimental considerations.

We demonstrate a novel method for electronically detecting and quantifying protein biomarkers using microfluidic impedance cytometry. Our biosensor, which consists of gold electrodes micro-fabricated in a microchannel, detects the differences between bead aggregates of varying sizes in a micro-pore sandwiched between two micro channels. We perform a sandwich immunoassay, where the complementary antibody pairs are immobilized on two different bead types, and the presence of antigen results in bead aggregation, the amount of which depends on antigen quantity. When single beads or bead aggregates pass through the impedance sensor, differences in impedance change are detected. In this manuscript, we perform a comprehensive theoretical study on the limits imposed on sensitivity of this technique due to electronic noise and also mass transfer and reaction limits. We also experimentally characterize the performance of this technique by validating the technique on an IgG detection assay. A detection limit at the picoMolar level is demonstrated, thus comparable in sensitivity to a sandwich ELISA.

Label-Free Single Microparticles and Cell Aggregates Sorting in Continuous Cell-Based Manufacturing.

There is a paradigm shift in biomanufacturing toward continuous bioprocessing but cell-based manufacturing using adherent and suspension cultures, including microcarriers, hydrogel microparticles, and 3D cell aggregates, remains challenging due to the lack of efficient in-line bioprocess monitoring and cell harvesting tools. Herein, a novel label-free microfluidic platform for high throughput (≈50 particles/sec) impedance bioanalysis of biomass, cell viability, and stem cell differentiation at single particle resolution is reported. The device is integrated with a real-time piezo-actuated particle sorter based on user-defined multi-frequency impedance signatures. Biomass profiling of Cytodex-3 microcarriers seeded with adipose-derived mesenchymal stem cells (ADSCs) is first performed to sort well-seeded or confluent microcarriers for downstream culture or harvesting, respectively. Next, impedance-based isolation of microcarriers with osteogenic differentiated ADSCs is demonstrated, which is validated with a twofold increase of calcium content in sorted ADSCs. Impedance profiling of heterogenous ADSCs-encapsulated hydrogel (alginate) microparticles and 3D ADSC aggregate mixtures is also performed to sort particles with high biomass and cell viability to improve cell quality. Overall, the scalable microfluidic platform technology enables in-line sample processing from bioreactors directly and automated analysis of cell quality attributes to maximize cell yield and improve the control of cell quality in continuous cell-based manufacturing.

Discrimination of Microplastics and Phytoplankton Using Impedance Cytometry.

Both microplastics and phytoplankton are found together in the ocean as suspended microparticles. There is a need for deployable technologies that can identify, size, and count these particles at high throughput to monitor plankton community structure and microplastic pollution levels. In situ analysis is particularly desirable as it avoids the problems associated with sample storage, processing, and degradation. Current technologies for phytoplankton and microplastic analysis are limited in their capability by specificity, throughput, or lack of deployability. Little attention has been paid to the smallest size fraction of microplastics and phytoplankton below 10 µm in diameter, which are in high abundance. Impedance cytometry is a technique that uses microfluidic chips with integrated microelectrodes to measure the electrical impedance of individual particles. Here, we present an impedance cytometer that can discriminate and count microplastics sampled directly from a mixture of phytoplankton in a seawater-like medium in the 1.5-10 µm size range. A simple machine learning algorithm was used to classify microplastic particles based on dual-frequency impedance measurements of particle size (at 1 MHz) and cell internal electrical composition (at 500 MHz). The technique shows promise for marine deployment, as the chip is sensitive, rugged, and mass producible.

Optimizing Microfluidic Impedance Cytometry by Bypass Electrode Layout Design.

Microfluidic impedance cytometry (MIC) has emerged as a popular technique for single-cell analysis. Traditional MIC electrode designs consist of a pair of (or three) working electrodes, and their detection performance needs further improvements for microorganisms. In this study, we designed an 8-electrode MIC device in which the center pair was defined as the working electrode, and the connection status of bypass electrodes could be changed. This allowed us to compare the performance of layouts with no bypasses and those with floating or grounding electrodes by simulation and experiment. The results of detecting Φ 5 µm beads revealed that both the grounding and the floating electrode outperformed the no bypass electrode, and the grounding electrode demonstrated the best signal-to-noise ratio (SNR), coefficient of variation (CV), and detection sensitivity. Furthermore, the effects of different bypass grounding areas (numbers of grounding electrodes) were investigated. Finally, particles passing at high horizontal positions can be detected, and Φ 1 µm beads can be measured in a wide channel (150 µm) using a fully grounding electrode, with the sensitivity of bead volume detection reaching 0.00097%. This provides a general MIC electrode optimization technology for detecting smaller particles, even macromolecular proteins, viruses, and exosomes in the future.

Phosphorylation-amplified synchronized droplet microfluidics sensitizes bacterial growth kinetic real-time monitoring.

The necessity for rapid and accurate bacterial growth monitoring is imperative across various domains, including healthcare and environmental safety. We introduce the self-synchronized droplet-amplified electrical screening cytometry (SYNC) system, a novel meld of droplet microfluidics and electrochemical amplification tailored for precise bacterial growth kinetic monitoring. SYNC encapsulates single bacteria in picolitre droplets, enabling real-time, fluorescence-free electrochemical monitoring. A specially devised phosphorylation-amplified culture medium translates bacterial metabolic activity into discernible electrical impedance changes. The dual-channel design and a rail-based structure in SYNC facilitate parallel screening and self-synchronization of droplets, addressing the limitations of conventional impedance cytometry. SYNC showcases a 5-fold enhancement in detection sensitivity and reduces 50% of the detection time compared to traditional approaches. Notably, SYNC is pioneering in providing exact initial bacterial concentrations, achieve to 104 bacteria/ml, a capability unmatched by existing real-time techniques measuring electrochemical variations. Along with its robust performance, this earmarks SYNC as a powerful tool for applications such as antibiotic susceptibility testing, food quality monitoring, and real-time water bacteria monitoring, paving the way for enhanced microbial process management and infection control.

Multi-modal sensing with integrated machine learning to differentiate specific leukocytes targeted by electrically sensitive hybrid particles.

The growing need for personalized, accurate, and non-invasive diagnostic technology has resulted in significant advancements, from pushing current mechanistic limitations to innovative modality developments across various disease-related biomarkers. However, there still lacks clinical solutions for analyzing multiple biomarkers simultaneously, limiting prognosis for patients suffering with complicated diseases or comorbidities. Here, we conceived, fabricated, and validated a multifrequency impedance cytometry apparatus with novel frequency-sensitive barcoded metal oxide Janus particles (MOJPs) as cell-receptor targeting agents. These microparticles are modulated by a metal oxide semi-coating which exhibit electrical property changes in a multifrequency electric field and are functionalized to target CD11b and CD66b membrane proteins on neutrophils. A multi-modal system utilizing supervised machine learning and simultaneous high-speed video microscopy classifies immune-specific surface receptors targeted by MOJPs as they form neutrophil-MOJP conjugates, based on multivariate multifrequency electrical recordings. High precision and sensitivity were determined based on the type of MOJPs conjugated with cells (>90% accuracy between neutrophil-MOJP conjugates versus cells alone). Remarkably, the design could differentiate the number of MOJPs conjugated per cell within the same MOJP class (>80% accuracy); which also improved comparing electrical responses across different MOJP types (>75% accuracy) as well. Such trends were consistent in individual blood samples and comparing consolidated data across multiple samples, demonstrating design robustness. The configuration may further expand to include more MOJP types targeting critical biomarker receptors in one sample and increase the modality's multiplexing potential.

Supervised learning on impedance cytometry data for label-free biophysical distinction of pancreatic cancer cells versus their associated fibroblasts under gemcitabine treatment.

Chemotherapy failure in pancreatic cancer patients is widely attributed to cancer cell reprogramming towards drug resistance by cancer associated fibroblasts (CAFs), which are the abundant cell type in the tumor microenvironment. Association of drug resistance to specific cancer cell phenotypes within multicellular tumors can advance isolation protocols for enabling cell-type specific gene expression markers to identify drug resistance. This requires the distinction of drug resistant cancer cells versus CAFs, which is challenging since permeabilization of CAF cells during drug treatment can cause non-specific uptake of cancer cell-specific stains. Cellular biophysical metrics, on the other hand, can provide multiparametric information to assess the gradual alteration of target cancer cells towards drug resistance, but these phenotypes need to be distinguished versus CAFs. Using pancreatic cancer cells and CAFs from a metastatic patient-derived tumor that exhibits cancer cell drug resistance under CAF co-culture, the biophysical metrics from multifrequency single-cell impedance cytometry are utilized for distinction of the subpopulation of viable cancer cells versus CAFs, before and after gemcitabine treatment. This is accomplished through supervised machine learning after training the model using key impedance metrics for cancer cells and CAFs from transwell co-cultures, so that an optimized classifier model can recognize each cell type and predict their respective proportions in multicellular tumor samples, before and after gemcitabine treatment, as validated by their confusion matrix and flow cytometry assays. In this manner, an aggregate of the distinguishing biophysical metrics of viable cancer cells after gemcitabine treatment in co-cultures with CAFs can be used in longitudinal studies, to classify and isolate the drug resistant subpopulation for identifying markers.

A Simple Micromilled Microfluidic Impedance Cytometer with Vertical Parallel Electrodes for Cell Viability Analysis.

Microfluidic impedance cytometry has been demonstrated as an effective platform for single cell analysis, taking advantage of microfabricated features and dielectric cell sensing methods. In this study, we present a simple microfluidic device to improve the sensitivity, accuracy, and throughput of single suspension cell viability analysis using vertical sidewall electrodes fabricated by a widely accessible negative manufacturing method. A microchannel milled through a 75 µm platinum wire, which was embedded into poly-methyl-methacrylate (PMMA), created a pair of parallel vertical sidewall platinum electrodes. Jurkat cells were interrogated in a custom low-conductivity buffer (1.2 ± 0.04 mS/cm) to reduce current leakage and increase device sensitivity. Confirmed by live/dead staining and electron microscopy, a single optimum excitation frequency of 2 MHz was identified at which live and dead cells were discriminated based on the disruption in the cell membrane associated with cell death. At this frequency, live cells were found to exhibit changes in the impedance phase with no appreciable change in magnitude, while dead cells displayed the opposite behavior. Correlated with video microscopy, a computational algorithm was created that could identify cell detection events and determine cell viability status by application of a mathematical correlation method.

Nucleic Acid Quantification by Multi-Frequency Impedance Cytometry and Machine Learning.

Determining nucleic acid concentrations in a sample is an important step prior to proceeding with downstream analysis in molecular diagnostics. Given the need for testing DNA amounts and its purity in many samples, including in samples with very small input DNA, there is utility of novel machine learning approaches for accurate and high-throughput DNA quantification. Here, we demonstrated the ability of a neural network to predict DNA amounts coupled to paramagnetic beads. To this end, a custom-made microfluidic chip is applied to detect DNA molecules bound to beads by measuring the impedance peak response (IPR) at multiple frequencies. We leveraged electrical measurements including the frequency and imaginary and real parts of the peak intensity within a microfluidic channel as the input of deep learning models to predict DNA concentration. Specifically, 10 different deep learning architectures are examined. The results of the proposed regression model indicate that an R_Squared of 97% with a slope of 0.68 is achievable. Consequently, machine learning models can be a suitable, fast, and accurate method to measure nucleic acid concentration in a sample. The results presented in this study demonstrate the ability of the proposed neural network to use the information embedded in raw impedance data to predict the amount of DNA concentration.