Deep learning detector for high precision monitoring of cell encapsulation statistics in microfluidic droplets

Machine learning-integrated droplet microfluidic system for accurate quantification and classification of microplastics

Microplastic (MP) pollution poses serious environmental and public health concerns, requiring efficient detection methods. Conventional techniques have the limitations of labor-intensive workflows and complex instrumentation, hindering rapid on-site field analysis. Here, we present the Machine learning (ML)-Integrated Droplet-based REal-time Analysis of MP (MiDREAM) system. Utilizing a compact peristaltic pump, the system achieved high-throughput droplet generation (> 200 Hz) while encapsulating MPs in uniform droplets (142 ± 8 μm). A high-resolution complementary metal oxide semiconductor (CMOS) sensor combined with an optimized YOLO v8 ML model was employed for real-time analysis, achieving a mean average precision (mAP) of 0.982 and an area under the curve (AUC) of 97.64 %. Comparative analysis with hemocytometer counting and surface-enhanced Raman spectroscopy (SERS) demonstrated the superior performance of the system, demonstrating high correlation (R² = 0.9965) and minimal deviation (6.36 %) from theoretical values. The system accurately classified MPs of different sizes, achieving accuracies of 95.4 %, 87.9 %, 95.3 %, 85.3 %, and 92.5 % for 3, 5, 10, 30, and 50 μm particles, respectively. Validation with real-world water samples confirmed the system adaptability, while maintaining high detection accuracy (> 90 %). The on-site field tests of MiDREAM system also demonstrated its robust performance for environmental monitoring in a variety of environments. Therefore, our portable and integrated MiDREAM system offers a promising solution for real-time environmental monitoring applications.

SAM-dPCR: Accurate and Generalist Nuclei Acid Quantification Leveraging the Zero-Shot Segment Anything Model

Digital PCR (dPCR) has transformed nucleic acid diagnostics by enabling the absolute quantification of rare mutations and target sequences. However, traditional dPCR detection methods, such as those involving flow cytometry and fluorescence imaging, may face challenges due to high costs, complexity, limited accuracy, and slow processing speeds. In this study, SAM-dPCR is introduced, a training-free open-source bioanalysis paradigm that offers swift and precise absolute quantification of biological samples. SAM-dPCR leverages the robustness of the zero-shot Segment Anything Model (SAM) to achieve rapid processing times (<4 seconds) with an accuracy exceeding 97.10%. This method has been extensively validated across diverse samples and reactor morphologies, demonstrating its broad applicability. Utilizing standard laboratory fluorescence microscopes, SAM-dPCR can measure nucleic acid template concentrations ranging from 0.154 copies µL-1 to 1.295 × 103 copies µL-1 for droplet dPCR and 0.160 × 103 to 3.629 × 103 copies µL-1 for microwell dPCR. Experimental validation shows a strong linear relationship (r2 > 0.96) between expected and determined sample concentrations. SAM-dPCR offers high accuracy, accessibility, and the ability to address bioanalytical needs in resource-limited settings, as it does not rely on hand-crafted "ground truth" data.

Deep Learning-Assisted Label-Free Parallel Cell Sorting with Digital Microfluidics

Sorting specific cells from heterogeneous samples is important for research and clinical applications. In this work, a novel label-free cell sorting method is presented that integrates deep learning image recognition with microfluidic manipulation to differentiate cells based on morphology. Using an Active-Matrix Digital Microfluidics (AM-DMF) platform, the YOLOv8 object detection model ensures precise droplet classification, and the Safe Interval Path Planning algorithm manages multi-target, collision-free droplet path planning. Simulations and experiments revealed that detection model precision, concentration ratios, and sorting cycles significantly affect recovery rates and purity. With HeLa cells and polystyrene beads as samples, the method achieved 98.5% sorting precision, 96.49% purity, and an 80% recovery over three cycles. After a series of experimental validations, this method can also be used to sort HeLa cells from red blood cells, cancer cells from white blood cells (represented by HeLa and Jurkat cells), and differentiate white blood cell subtypes (represented by HL-60 cells and Jurkat cells). Cells sorted using this method can be lysed directly on chip within their hosting droplets, ensuring minimal sample loss and suitability for downstream bioanalysis. This innovative AM-DMF cell sorting technique holds significant potential to advance diagnostics, therapeutics, and fundamental research in cell biology.

Pump-Free Microfluidics for Cell Concentration Analysis on Smartphones in Clinical Settings (SmartFlow): Design, Development, and Evaluation

BACKGROUND

Cell concentration in body fluid is an important factor for clinical diagnosis. The traditional method involves clinicians manually counting cells under microscopes, which is labor-intensive. Automated cell concentration estimation can be achieved using flow cytometers; however, their high cost limits accessibility. Microfluidic systems, although cheaper than flow cytometers, still require high-speed cameras and syringe pumps to drive the flow and ensure video quality. In this paper, we present SmartFlow, a low-cost solution for cell concentration estimation using smartphone-based computer vision on 3D-printed, pump-free microfluidic platforms.

OBJECTIVE

The objective was to design and fabricate microfluidic chips, coupled with clinical utilities, for cell counting and concentration analysis. We answered the following research questions (RQs): RQ1, Can gravity drive the flow within the microfluidic chips, eliminating the need for external pumps? RQ2, How does the microfluidic chip design impact video quality for cell analysis? RQ3, Can smartphone-captured videos be used to estimate cell count and concentration in microfluidic chips?

METHODS

To answer the 3 RQs, 2 experiments were conducted. In the cell flow velocity experiment, diluted sheep blood flowed through the microfluidic chips with and without a bottleneck design to answer RQ1 and RQ2, respectively. In the cell concentration analysis experiment, sheep blood diluted into 13 concentrations flowed through the microfluidic chips while videos were recorded by smartphones for the concentration measurement.

RESULTS

In the cell flow velocity experiment, we designed and fabricated 2 versions of microfluidic chips. The ANOVA test (Straight: F6, 99=6144.45, P<.001; Bottleneck: F6, 99=3475.78, P<.001) showed the height difference had a significant impact on the cell velocity, which implied gravity could drive the flow. The video sharpness analysis demonstrated that video quality followed an exponential decay with increasing height differences (video quality=100e-k×Height) and a bottleneck design could effectively preserve video quality (Straight: R2=0.95, k=4.33; Bottleneck: R2=0.91, k=0.59). Samples from the 13 cell concentrations were used for cell counting and cell concentration estimation analysis. The accuracy of cell counting (n=35, 60-second samples, R2=0.96, mean absolute error=1.10, mean squared error=2.24, root mean squared error=1.50) and cell concentration regression (n=39, 150-second samples, R2=0.99, mean absolute error=0.24, mean squared error=0.11, root mean squared error=0.33 on a logarithmic scale, mean average percentage error=0.25) were evaluated using 5-fold cross-validation by comparing the algorithmic estimation to ground truth.

CONCLUSIONS

In conclusion, we demonstrated the importance of the flow velocity in a microfluidic system, and we proposed SmartFlow, a low-cost system for computer vision-based cellular analysis. The proposed system could count the cells and estimate cell concentrations in the samples.

Data-driven models for microfluidics: A short review

Microfluidic devices have many unique practical applications across a wide range of fields, making it important to develop accurate models of these devices, and many different models have been developed. Existing modeling methods mainly include mechanism derivation and semi-empirical correlations, but both are not universally applicable. In order to achieve a more accurate and general modeling process, the use of data-driven modeling has been studied recently. This review highlights recent advances in the application of data-driven modeling techniques for simulating and designing microfluidic devices. First, it introduces the application of traditional modeling approaches in microfluidics; subsequently, through different database sources, it reviews studies on data-driven modeling in three categories; and finally, it raises some open issues that require further investigation.

Rapid and portable quantification of HIV RNA via a smartphone-enabled digital CRISPR device and deep learning

For the 29.8 million people in the world living with HIV/AIDS and receiving antiretroviral therapy, it is crucial to monitor their HIV viral loads. To this end, rapid and portable diagnostic tools that can quantify HIV RNA are critically needed. We report herein a rapid and quantitative digital CRISPR-assisted HIV RNA detection assay that has been implemented within a portable smartphone-based device as a potential solution. Specifically, we first developed a fluorescence-based reverse transcription recombinase polymerase amplification (RT-RPA)-CRISPR assay that can efficiently detect HIV RNA at 42 °C. We then implemented this assay within a commercial stamp-sized digital chip, where RNA molecules were quantified as strongly fluorescent digital reaction wells. The isothermal reaction condition and the strong fluorescence in the digital chip simplified the design of thermal and optical modules, allowing us to engineer a palm-size device measuring 70 × 115 × 80 mm and weighing less than 0.6 kg. We also capitalized the smartphone by developing a custom app to control the device, perform the digital assay, and capture fluorescence images throughout the assay using the smartphone's camera. Moreover, we trained and verified a deep learning-based algorithm for analyzing fluorescence images and identifying positive digital reaction wells with high accuracy. Using our smartphone-enabled digital CRISPR device, we successfully detected as low as 75 copies of HIV RNA in just 15 min, showing its potential toward monitoring of HIV viral loads and aiding the global effort to combat the HIV/AIDS epidemic.

Artificial intelligence-enabled multipurpose smart detection in active-matrix electrowetting-on-dielectric digital microfluidics

An active-matrix electrowetting-on-dielectric (AM-EWOD) system integrates hundreds of thousands of active electrodes for sample droplet manipulation, which can enable simultaneous, automatic, and parallel on-chip biochemical reactions. A smart detection system is essential for ensuring a fully automatic workflow and online programming for the subsequent experimental steps. In this work, we demonstrated an artificial intelligence (AI)-enabled multipurpose smart detection method in an AM-EWOD system for different tasks. We employed the U-Net model to quantitatively evaluate the uniformity of the applied droplet-splitting methods. We used the YOLOv8 model to monitor the droplet-splitting process online. A 97.76% splitting success rate was observed with 18 different AM-EWOD chips. A 99.982% model precision rate and a 99.980% model recall rate were manually verified. We employed an improved YOLOv8 model to detect single-cell samples in nanolitre droplets. Compared with manual verification, the model achieved 99.260% and 99.193% precision and recall rates, respectively. In addition, single-cell droplet sorting and routing experiments were demonstrated. With an AI-based smart detection system, AM-EWOD has shown great potential for use as a ubiquitous platform for implementing true lab-on-a-chip applications.

AI-Powered Microfluidics: Shaping the Future of Phenotypic Drug Discovery

Phenotypic drug discovery (PDD), which involves harnessing biological systems directly to uncover effective drugs, has undergone a resurgence in recent years. The rapid advancement of artificial intelligence (AI) over the past few years presents numerous opportunities for augmenting phenotypic drug screening on microfluidic platforms, leveraging its predictive capabilities, data analysis, efficient data processing, etc. Microfluidics coupled with AI is poised to revolutionize the landscape of phenotypic drug discovery. By integrating advanced microfluidic platforms with AI algorithms, researchers can rapidly screen large libraries of compounds, identify novel drug candidates, and elucidate complex biological pathways with unprecedented speed and efficiency. This review provides an overview of recent advances and challenges in AI-based microfluidics and their applications in drug discovery. We discuss the synergistic combination of microfluidic systems for high-throughput screening and AI-driven analysis for phenotype characterization, drug-target interactions, and predictive modeling. In addition, we highlight the potential of AI-powered microfluidics to achieve an automated drug screening system. Overall, AI-powered microfluidics represents a promising approach to shaping the future of phenotypic drug discovery by enabling rapid, cost-effective, and accurate identification of therapeutically relevant compounds.

Cellular nucleus image-based smarter microscope system for single cell analysis

Cell imaging technology is undoubtedly a powerful tool for studying single-cell heterogeneity due to its non-invasive and visual advantages. It covers microscope hardware, software, and image analysis techniques, which are hindered by low throughput owing to abundant hands-on time and expertise. Herein, a cellular nucleus image-based smarter microscope system for single-cell analysis is reported to achieve high-throughput analysis and high-content detection of cells. By combining the hardware of an automatic fluorescence microscope and multi-object recognition/acquisition software, we have achieved more advanced process automation with the assistance of Robotic Process Automation (RPA), which realizes a high-throughput collection of single-cell images. Automated acquisition of single-cell images has benefits beyond ease and throughout and can lead to uniform standard and higher quality images. We further constructed a single-cell image database-based convolutional neural network (Efficient Convolutional Neural Network, E-CNN) exceeding 20618 single-cell nucleus images. Computational analysis of large and complex data sets enhances the content and efficiency of single-cell analysis with the assistance of Artificial Intelligence (AI), which breaks through the super-resolution microscope's hardware limitation, such as specialized light sources with specific wavelengths, advanced optical components, and high-performance graphics cards. Our system can identify single-cell nucleus images that cannot be artificially distinguished with an accuracy of 95.3%. Overall, we build an ordinary microscope into a high-throughput analysis and high-content smarter microscope system, making it a candidate tool for Imaging cytology.

Dynamic video recognition for cell-encapsulating microfluidic droplets

Droplet microfluidics is a highly sensitive and high-throughput technology extensively utilized in biomedical applications, such as single-cell sequencing and cell screening. However, its performance is highly influenced by the droplet size and single-cell encapsulation rate (following random distribution), thereby creating an urgent need for quality control. Machine learning has the potential to revolutionize droplet microfluidics, but it requires tedious pixel-level annotation for network training. This paper investigates the application software of the weakly supervised cell-counting network (WSCApp) for video recognition of microdroplets. We demonstrated its real-time performance in video processing of microfluidic droplets and further identified the locations of droplets and encapsulated cells. We verified our methods on droplets encapsulating six types of cells/beads, which were collected from various microfluidic structures. Quantitative experimental results showed that our approach can not only accurately distinguish droplet encapsulations (micro-F1 score > 0.94), but also locate each cell without any supervised location information. Furthermore, fine-tuning transfer learning on the pre-trained model also significantly reduced (>80%) annotation. This software provides a user-friendly and assistive annotation platform for the quantitative assessment of cell-encapsulating microfluidic droplets.

High-throughput microfluidic systems accelerated by artificial intelligence for biomedical applications

High-throughput microfluidic systems are widely used in biomedical fields for tasks like disease detection, drug testing, and material discovery. Despite the great advances in automation and throughput, the large amounts of data generated by the high-throughput microfluidic systems generally outpace the abilities of manual analysis. Recently, the convergence of microfluidic systems and artificial intelligence (AI) has been promising in solving the issue by significantly accelerating the process of data analysis as well as improving the capability of intelligent decision. This review offers a comprehensive introduction on AI methods and outlines the current advances of high-throughput microfluidic systems accelerated by AI, covering biomedical detection, drug screening, and automated system control and design. Furthermore, the challenges and opportunities in this field are critically discussed as well.

Development and future of droplet microfluidics

Over the past two decades, advances in droplet-based microfluidics have facilitated new approaches to process and analyze samples with unprecedented levels of precision and throughput. A wide variety of applications has been inspired across multiple disciplines ranging from materials science to biology. Understanding the dynamics of droplets enables optimization of microfluidic operations and design of new techniques tailored to emerging demands. In this review, we discuss the underlying physics behind high-throughput generation and manipulation of droplets. We also summarize the applications in droplet-derived materials and droplet-based lab-on-a-chip biotechnology. In addition, we offer perspectives on future directions to realize wider use of droplet microfluidics in industrial production and biomedical analyses.

Microfluidics-aided fabrication of 3D micro-nano hierarchical SERS substrate for rapid detection of dual hepatocellular carcinoma biomarkers

The simultaneous analysis of trace amounts of dual biomarkers is crucial in the early diagnosis, treatment, and prognosis of hepatocellular carcinoma (HCC). In this study, we prepared SERS-active hydrogel microparticles (SAHMs) with 3D hierarchical gold nanoparticles (AuNPs) micro-nanostructures by microdroplet technology and in situ synthesis, which demonstrated high reproducibility and sensitivity. Compared with traditional 2D SERS substrates, this newly prepared 3D SERS substrate provided a high density of nano-wrinkled structures and numerous AuNPs. Furthermore, a newly designed SERS-active substrate was proposed for the simultaneous microfluidic detection of AFP and AFU. The Raman signals of sandwich immunocomplexes on the surface of the SAHMs were measured for the trace analysis of these biomarkers. The proposed microfluidic platform achieved AFP and AFU detection in the range of 0.1-100 ng mL-1 and 0.01-100 ng mL-1, respectively, which represents a good response. Indeed, this platform is easy to fabricate, of low cost and has short detection time and comparable detection limits to other methods. As far as we know, this is the first study to achieve the simultaneous detection of AFP and AFU on a microfluidic platform. Therefore, we proposed a new simultaneous detection platform for dual HCC biomarkers that shows strong potential for the early diagnosis of HCC.

Image-Based Feedback of Multi-Component Microdroplets for Ultra-Monodispersed Library Preparation

Using devices with microfluidic channels can allow for precise control over liquids flowing through them. Merging flows of immiscible liquids can create emulsions with highly monodispersed microdroplets within a carrier liquid, which are ideal for miniaturised reaction vessels which can be generated with a high throughput of tens of thousands of droplets per second. Control of the size and composition of these droplets is generally performed by controlling the pumping system pushing the liquids into the device; however, this is an indirect manipulation and inadequate if absolute precision is required in the size or composition of the droplets. In this work, we extend the previous development of image-based closed-loop feedback control over microdroplet generation to allow for the control of not only the size of droplets but also the composition by merging two aqueous flows. The feedback allows direct control over the desired parameters of volume and ratio of the two components over a wide range of ratios and outperforms current techniques in terms of monodispersity in volume and composition. This technique is ideal for situations where precise control over droplets is critical, or where a library of droplets of different concentrations but the same volume is required.

A general neural network model co-driven by mechanism and data for the reliable design of gas-liquid T-junction microdevices

In recent years, many models have been developed to describe the gas-liquid microdispersion process, which mainly rely on mechanistic analysis and may not be universally applicable. In order to provide a more comprehensive model and, most significantly, to provide a model for design, we have established a general database of microbubble generation in T-junction microdevices, including 854 data points from 12 pieces of literature. A neural network model that combines mechanistic and data modeling is developed. By transfer learning, more accurate results can be obtained. Additionally, we have proposed a design method that enables a relative deviation of less than 5% from the expected bubble size. A new device was designed and prepared to confirm the reliability of the method, which can prepare smaller bubbles than other common T-junction devices. In this way, a general and universal database and model are established and a design method for a gas-liquid T-junction microreactor is developed.

Towards single cell encapsulation for precision biology and medicine

The primary impetus of therapeutic cell encapsulation in the past several decades has been to broaden the options for donor cell sources by countering against immune-mediated rejection. However, another significant advantage of encapsulation is to provide donor cells with physiologically relevant cues that become compromised in disease. The advances in biomaterial design have led to the fundamental insight that cells sense and respond to various signals encoded in materials, ranging from biochemical to mechanical cues. The biomaterial design for cell encapsulation is becoming more sophisticated in controlling specific aspects of cellular phenotypes and more precise down to the single cell level. This recent progress offers a paradigm shift by designing single cell-encapsulating materials with predefined cues to precisely control donor cells after transplantation.

Computer vision meets microfluidics: a label-free method for high-throughput cell analysis

In this paper, we review the integration of microfluidic chips and computer vision, which has great potential to advance research in the life sciences and biology, particularly in the analysis of cell imaging data. Microfluidic chips enable the generation of large amounts of visual data at the single-cell level, while computer vision techniques can rapidly process and analyze these data to extract valuable information about cellular health and function. One of the key advantages of this integrative approach is that it allows for noninvasive and low-damage cellular characterization, which is important for studying delicate or fragile microbial cells. The use of microfluidic chips provides a highly controlled environment for cell growth and manipulation, minimizes experimental variability and improves the accuracy of data analysis. Computer vision can be used to recognize and analyze target species within heterogeneous microbial populations, which is important for understanding the physiological status of cells in complex biological systems. As hardware and artificial intelligence algorithms continue to improve, computer vision is expected to become an increasingly powerful tool for in situ cell analysis. The use of microelectromechanical devices in combination with microfluidic chips and computer vision could enable the development of label-free, automatic, low-cost, and fast cellular information recognition and the high-throughput analysis of cellular responses to different compounds, for broad applications in fields such as drug discovery, diagnostics, and personalized medicine.

WSCNet: Biomedical Image Recognition for Cell Encapsulated Microfluidic Droplets

Microfluidic droplets accommodating a single cell as independent microreactors are frequently demanded for single-cell analysis of phenotype and genotype. However, challenges exist in identifying and reducing the covalence probability (following Poisson's distribution) of more than two cells encapsulated in one droplet. It is of great significance to monitor and control the quantity of encapsulated content inside each droplet. We demonstrated a microfluidic system embedded with a weakly supervised cell counting network (WSCNet) to generate microfluidic droplets, evaluate their quality, and further recognize the locations of encapsulated cells. Here, we systematically verified our approach using encapsulated droplets from three different microfluidic structures. Quantitative experimental results showed that our approach can not only distinguish droplet encapsulations (F1 score > 0.88) but also locate each cell without any supervised location information (accuracy > 89%). The probability of a "single cell in one droplet" encapsulation is systematically verified under different parameters, which shows good agreement with the distribution of the passive method (Residual Sum of Squares, RSS < 0.5). This study offers a comprehensive platform for the quantitative assessment of encapsulated microfluidic droplets.

Deep learning with microfluidics for on-chip droplet generation, control, and analysis

Droplet microfluidics has gained widespread attention in recent years due to its advantages of high throughput, high integration, high sensitivity and low power consumption in droplet-based micro-reaction. Meanwhile, with the rapid development of computer technology over the past decade, deep learning architectures have been able to process vast amounts of data from various research fields. Nowadays, interdisciplinarity plays an increasingly important role in modern research, and deep learning has contributed greatly to the advancement of many professions. Consequently, intelligent microfluidics has emerged as the times require, and possesses broad prospects in the development of automated and intelligent devices for integrating the merits of microfluidic technology and artificial intelligence. In this article, we provide a general review of the evolution of intelligent microfluidics and some applications related to deep learning, mainly in droplet generation, control, and analysis. We also present the challenges and emerging opportunities in this field.

Blood quality evaluation via on-chip classification of cell morphology using a deep learning algorithm

The quality of red blood cells (RBCs) in stored blood has a direct impact on the recovery of patients treated by blood transfusion, which directly reflects the quality of blood. The traditional means for blood quality evaluation involve the use of reagents and multi-step and time-consuming operations. Here, a low-cost, multi-classification, label-free and high-precision method is developed, which combines microfluidic technology and a deep learning algorithm together to recognize and classify RBCs based on morphology. The microfluidic channel is designed to effectively and controllably solve the problem of cell overlap, which has a severe negative impact on the identification of cells. The object detection model in the deep learning algorithm is optimized and used to recognize multiple RBCs simultaneously in the whole field of view, so as to classify them into six morphological subcategories and count the numbers in each subgroup. The mean average precision of the developed object detection model reaches 89.24%. The blood quality can be evaluated by calculating the morphology index (MI) according to the numbers of cells in subgroups. The validation of the method is verified by evaluating three blood samples stored for 7 days, 21 days and 42 days, which have MIs of 84.53%, 73.33% and 24.34%, respectively, indicating good agreement with the actual blood quality. This method has the merits of cell identification in a wide channel, no need for single cell alignment as the image cytometry does and it is not only applicable to the quality evaluation of RBCs, but can also be used for general cell identifications with different morphologies.

Microsystem Advances through Integration with Artificial Intelligence

Microfluidics is a rapidly growing discipline that involves studying and manipulating fluids at reduced length scale and volume, typically on the scale of micro- or nanoliters. Under the reduced length scale and larger surface-to-volume ratio, advantages of low reagent consumption, faster reaction kinetics, and more compact systems are evident in microfluidics. However, miniaturization of microfluidic chips and systems introduces challenges of stricter tolerances in designing and controlling them for interdisciplinary applications. Recent advances in artificial intelligence (AI) have brought innovation to microfluidics from design, simulation, automation, and optimization to bioanalysis and data analytics. In microfluidics, the Navier-Stokes equations, which are partial differential equations describing viscous fluid motion that in complete form are known to not have a general analytical solution, can be simplified and have fair performance through numerical approximation due to low inertia and laminar flow. Approximation using neural networks trained by rules of physical knowledge introduces a new possibility to predict the physicochemical nature. The combination of microfluidics and automation can produce large amounts of data, where features and patterns that are difficult to discern by a human can be extracted by machine learning. Therefore, integration with AI introduces the potential to revolutionize the microfluidic workflow by enabling the precision control and automation of data analysis. Deployment of smart microfluidics may be tremendously beneficial in various applications in the future, including high-throughput drug discovery, rapid point-of-care-testing (POCT), and personalized medicine. In this review, we summarize key microfluidic advances integrated with AI and discuss the outlook and possibilities of combining AI and microfluidics.