Smartphone-based platforms implementing microfluidic detection with image-based artificial intelligence

Rapid parallel blood typing on centrifugal microfluidic platform by microcolumn gel immunoassay

Microcolumn gel immunoassay (MGIA) has the ability to meet the requirements of clinical diagnosis due to its reliable sensitivity and accuracy. However, traditional MGIA exhibits limitations including inadequate portability, low throughput, and extended analysis time. To address these challenges, we combined MGIA with microfluidic technology, demonstrating a centrifugal microfluidic-based microcolumn gel immunoassay (μMGIA) platform for blood typing of clinical samples. Experimental results indicate that the μMGIA platform can simultaneously detect six blood group antigens in five clinical blood samples within 2 min. Notably, it offers comprehensive detection of ABO blood group antigens and Rh blood group antigens with 100 % accuracy, outperforming the traditional slide method. The integration of microfluidic technology with MGIA circumvents the constraints of traditional methods, providing a new avenue for blood typing and immunoanalysis of clinical samples.

Machine-Learning Mental-Fatigue-Measuring μm-Thick Elastic Epidermal Electronics (MMMEEE)

Electrophysiological (EP) signals are key biomarkers for monitoring mental fatigue (MF) and general health, but state-of-the-art wearable EP-based MF monitoring systems are bulky and require user-specific, labeled data. Ultrathin epidermal electrodes with high performance are ideal for constructing imperceptive EP sensing systems; however, the lack of a simple and scalable fabrication delays their application in MF recognition. Here, we report a facile, scalable printing-welding-transferring strategy (PWT) for printing μm-thickness micropatterned silver nanowires (AgNWs)/sticky polydimethylsiloxane, welding the AgNWs via plasmonic effect, and transferring the electrode to skin as tattoos. The PWT provides electrodes with conformability, comfort, and stability for EP sensing. Leveraging the facile and scalable PWT, we develop plug-and-play wireless multimodal epidermal electronics integrated with an unsupervised transfer learning (UTL) scheme for MF recognition across various users. The UTL adaptively minimizes the intersubject difference and achieves high accuracy, without demand of expensive computation and labels from target users.

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.

AI-empowered visualization of nucleic acid testing

AIMS

The visualization of nucleic acid testing (NAT) results plays a critical role in diagnosing and monitoring infectious and genetic diseases. The review aims to review the current status of AI-based NAT result visualization. It systematically introduces commonly used AI-based methods and techniques for NAT, emphasizing the importance of result visualization for accessible, clear, and rapid interpretation. This highlights the importance of developing a NAT visualization platform that is user-friendly and efficient, setting a clear direction for future advancements in making nucleic acid testing more accessible and effective for everyday applications.

METHOD

This review explores both the commonly used NAT methods and AI-based techniques for NAT result visualization. The focus then shifts to AI-based methodologies, such as color detection and result interpretation through AI algorithms. The article presents the advantages and disadvantages of these techniques, while also comparing the performance of various NAT platforms in different experimental contexts. Furthermore, it explores the role of AI in enhancing the accuracy, speed, and user accessibility of NAT results, highlighting visualization technologies adapted from other fields of experimentation.

SIGNIFICANCE

This review offers valuable insights for researchers and everyday users, aiming to develop effective visualization platforms for NAT, ultimately enhancing disease diagnosis and monitoring.

Transformative laboratory medicine enabled by microfluidic automation and artificial intelligence

Laboratory medicine provides pivotal medical information through analyses of body fluids and tissues, and thus, it is essential for diagnosis of diseases as well as monitoring of disease progression. Despite its universal importance, the field is currently suffering from the limited workforce and analytical capabilities due to the increasing pressure from expanding global population and unexpected rise of noncommunicable diseases. The emerging technologies of microfluidic automation and artificial intelligence (AI) has led to the development of advanced diagnostic platforms, positioning themselves as adaptable solutions to enable highly efficient and accessible laboratory medicine. In this review, we will provide a comprehensive review of microfluidic automation, focusing on the microstructure design and automation principles, along with its intended functionalities for diagnostic purposes. Subsequently, we exemplify the integration of AI with microfluidics and illustrating how their combination benefits for the applications and what the challenges are in this rapidly evolving field. Finally, the review offers a balanced perspective on the microfluidics and AI, discussing their promising role in advancing laboratory medicine.

A press-actuated slidable microfluidic colorimetric biosensor for Salmonella detection utilizing nickel mesh sheet and MIL-88@Pd/Pt nanoparticles

A press-actuated slidable microfluidic colorimetric biosensor was designed for rapid, sensitive and multi-channel detection of Salmonella. The nickel mesh sheet (NMS) modified with capture antibodies was employed for capturing target bacteria, and metal organic frameworks decorated with palladium (Pd) and platinum (Pt) nanoparticles (MIL-88@Pd/Pt NPs) modified with detection antibodies were used for amplifying colorimetric signals. The capture efficiency of the immune NMS reached 83 %, and the detection limit of this colorimetric biosensor was 35 CFU/mL in 20 min. The average recoveries for Salmonella in spiked chicken meats ranged from 92.2 % to 102.5 % with a variation from 3.7 % to 7.2 %. The press-actuated slidable microfluidic chip was elaboratively developed with multiple functions, including mixing, separation, washing, catalysis and detection.

A microfluidic chip with integrated plasma separation for sample-to-answer detection of multiple chronic disease biomarkers in whole blood

Point-of-care testing of multiple chronic disease biomarkers is crucial for timely intervention and management of chronic diseases. Here, a "sample-to-answer" microfluidic chip was developed for simultaneous detection of multiple chronic disease biomarkers in whole blood by integrating a plasma separation module. The whole detection process is very convenient, i.e., just add whole blood and get the results. The chip successfully achieved the simultaneous detection of total cholesterol, triglycerides, uric acid, and glucose in undiluted whole blood within 21 min, including 6 min for plasma separation and 15 min for enzymatic chromogenic reactions. Moreover, the sensitivity levels of on-chip detection of chronic disease biomarkers can also meet clinically relevant thresholds. The chip is easy to use and has significant potential to improve home self-management of chronic diseases and enhance healthcare outcomes.

Frog tongue-inspired wettable microfibers for particles capture

Fibers have been of great significance in our daily lives, especially in the industrial production of masks. Research in this area has been focused on developing microfibers with superior functions to enhance the filtration performances of the masks. Herein, inspired by the frog's predation mechanism using its tongues to swiftly grab flying insects, we propose novel porous wettable microfibers from microfluidics to efficiently capture particles in the air for filtration. Upon pre-dispersing LP emulsions into polyurethane (PU), porous microfibers dispersed with oil droplets could be continuously spun from a co-flow microfluidic device based on the quick phase inversion of PU. To design an optimal system with frog-tongue-like interfacial adhesion properties, the wettability performances of the porous microfibers are investigated under full, partial, and no oil coverage conditions. When implemented in a mask, the 3D patterned networks based on the frog-tongue-inspired microfibers have been proven with remarkable particle capture performances while maintaining good air permeability. Based on these features, we believe that frog-tongue-inspired microfibers and their derived masks are of practical significance in multiple applications.

An electrochemical point-of-care testing device for specific diagnosis of the albinism biomarker based on paradigm shift designs

L-tyrosine is a recognized biomarker of albinism, whose endogenous level in human bodies is directly linked to melanin synthesis while no attention has been paid to its specific diagnosis. To this end, we have developed an electrochemical point-of-care testing device based on a molecularly imprinted gel prepared by a universal paradigm shift design to achieve the enhanced specific recognition of the L-tyrosine. Interestingly, this theoretically optimized molecularly imprinted gel validates the recognition pattern of L-tyrosine and optimizes the structure of the polymer itself with the aid of computational chemistry. Besides, modified extended-layer MXene and Au nanoclusters have significantly improved the sensing activity. As a result, the linear diagnostic range of this electrochemical point-of-care testing device for L-tyrosine is 0.1-100 μM in actual human fluids, which fully covers the L-tyrosine levels of healthy individuals and people with albinism. The diagnosis is completed in 90 s and then the results are transmitted by Bluetooth low energy to the smart mobile terminal. Therefore, we are convinced that this electrochemical point-of-care testing device is a promising tool in the future smart medical system.

Emerging Trends in Integrated Digital Microfluidic Platforms for Next-Generation Immunoassays

Technologies based on digital microfluidics (DMF) have made significant advancements in the automated manipulation of microscale liquids and complex multistep processes. Due to their numerous benefits, such as automation, speed, cost-effectiveness, and minimal sample volume requirements, these systems are particularly well suited for immunoassays. In this review, an overview is provided of diverse DMF manipulation platforms and their applications in immunological analysis. Initially, droplet-driven DMF platforms based on electrowetting on dielectric (EWOD), magnetic manipulation, surface acoustic wave (SAW), and other related technologies are briefly introduced. The preparation of DMF is then described, including material selection, fabrication techniques and droplet generation. Subsequently, a comprehensive account of advancements in the integration of DMF with various immunoassay techniques is offered, encompassing colorimetric, direct chemiluminescence, enzymatic chemiluminescence, electrosensory, and other immunoassays. Ultimately, the potential challenges and future perspectives in this burgeoning field are delved into.

Phase Transition of Wax Enabling CRISPR Diagnostics for Automatic At-Home Testing of Multiple Sexually Transmitted Infection Pathogens

Sexually transmitted infections (STIs) significantly impact women's reproductive health. Rapid, sensitive, and affordable detection of these pathogens is essential, especially for home-based self-testing, which is crucial for individuals who prioritize privacy or live in areas with limited access to healthcare services. Herein, an automated diagnostic system called Wax-CRISPR has been designed specifically for at-home testing of multiple STIs. This system employs a unique strategy by using the solid-to-liquid phase transition of wax to sequentially isolate and mix recombinase polymerase amplification (RPA) and CRISPR assays in a microfluidic chip. By incorporating a home-built controlling system, Wax-CRISPR achieves true one-pot multiplexed detection. The system can simultaneously detect six common critical gynecological pathogens (CT, MG, UU, NG, HPV 16, and HPV 18) within 30 min, with a detection limit reaching 10-18 M. Clinical evaluation demonstrates that the system achieves a sensitivity of 96.8% and a specificity of 97.3% across 100 clinical samples. Importantly, eight randomly recruited untrained operators performe a double-blinded test and successfully identified the STI targets in 33 clinical samples. This wax-transition-based one-pot CRISPR assay offers advantages such as low-cost, high-stability, and user-friendliness, making it a useful platform for at-home or field-based testing of multiple pathogen infections.

Smartphone-driven centrifugal microfluidics for diagnostics in resource limited settings

The broad availability of smartphones has provided new opportunities to develop less expensive, portable, and integrated point-of-care (POC) platforms. Here, a platform that consists of three main components is introduced: a portable housing, a centrifugal microfluidic disc, and a mobile phone. The mobile phone supplies the electrical power and serves as an analysing system. The low-cost housing made from cardboard serves as a platform to conduct tests. The electrical energy stored in mobile phones was demonstrated to be adequate for spinning a centrifugal disc up to 3000 revolutions per minute (RPM), a rotation speed suitable for majority of centrifugal microfluidics-based assays. For controlling the rotational speed, a combination of magnetic and acoustic tachometry using embedded sensors of the mobile phone was used. Experimentally, the smartphone-based tachometry was proven to be comparable with a standard laser-based tachometer. As a proof of concept, two applications were demonstrated using the portable platform: a colorimetric sandwich immunoassay to detect interleukin-2 (IL-2) having a limit of detection (LOD) of 65.17 ng/mL and a fully automated measurement of hematocrit level integrating blood-plasma separation, imaging, and image analysis that takes less than 5 mins to complete. The low-cost platform weighing less than 150 g and operated by a mobile phone has the potential to meet the REASSURED criteria for advanced diagnostics in resource limited settings.

Inclusive and Accurate Clinical Diagnostics Using Intelligent Computation and Smartphone Imaging

Smartphone-based colorimetry has been widely applied in clinical analysis, although significant challenges remain in its practical implementation, including the need to consider biases introduced by the ambient imaging environment, which limit its potential within a clinical decision pathway. In addition, most commercial devices demonstrate variability introduced by manufacturer-to-manufacturer differences. Here, we undertake a systematic characterization of the potential imaging interferences that lead to this limited performance in conventional smartphones and, in doing so, provide a comprehensive new understanding of smartphone color imaging. Through derivation of a strongly correlated parameter for sample quantification, we enable real-time imaging, which for the first time, takes the first steps to turning the mobile phone camera into an analytical instrument - irrespective of model, software, and the operating systems used. We demonstrate clinical applicability through the imaging of patients' skin, enabling rapid and convenient diagnosis of cyanosis and measurement of local oxygen concentration to a level that unlocks clinical decision-making for monitoring cardiovascular disease and anemia. Importantly, we show that our solution also accounts for the differences in individuals' skin tones as measured across the Fitzpatrick scale, overcoming potential clinically significant errors in current optical oximetry.

Artificial intelligence performance in testing microfluidics for point-of-care

Artificial intelligence (AI) is revolutionizing medicine by automating tasks like image segmentation and pattern recognition. These AI approaches support seamless integration with existing platforms, enhancing diagnostics, treatment, and patient care. While recent advancements have demonstrated AI superiority in advancing microfluidics for point of care (POC) diagnostics, a gap remains in comparative evaluations of AI algorithms in testing microfluidics. We conducted a comparative evaluation of AI models specifically for the two-class classification problem of identifying the presence or absence of bubbles in microfluidic channels under various imaging conditions. Using a model microfluidic system with a single channel loaded with 3D transparent objects (bubbles), we challenged each of the tested machine learning (ML) (n = 6) and deep learning (DL) (n = 9) models across different background settings. Evaluation revealed that the random forest ML model achieved 95.52% sensitivity, 82.57% specificity, and 97% AUC, outperforming other ML algorithms. Among DL models suitable for mobile integration, DenseNet169 demonstrated superior performance, achieving 92.63% sensitivity, 92.22% specificity, and 92% AUC. Remarkably, DenseNet169 integration into a mobile POC system demonstrated exceptional accuracy (>0.84) in testing microfluidics at under challenging imaging settings. Our study confirms the transformative potential of AI in healthcare, emphasizing its capacity to revolutionize precision medicine through accurate and accessible diagnostics. The integration of AI into healthcare systems holds promise for enhancing patient outcomes and streamlining healthcare delivery.

Advances in Portable Optical Microscopy Using Cloud Technologies and Artificial Intelligence for Medical Applications

The need for faster and more accessible alternatives to laboratory microscopy is driving many innovations throughout the image and data acquisition chain in the biomedical field. Benchtop microscopes are bulky, lack communications capabilities, and require trained personnel for analysis. New technologies, such as compact 3D-printed devices integrated with the Internet of Things (IoT) for data sharing and cloud computing, as well as automated image processing using deep learning algorithms, can address these limitations and enhance the conventional imaging workflow. This review reports on recent advancements in microscope miniaturization, with a focus on emerging technologies such as photoacoustic microscopy and more established approaches like smartphone-based microscopy. The potential applications of IoT in microscopy are examined in detail. Furthermore, this review discusses the evolution of image processing in microscopy, transitioning from traditional to deep learning methods that facilitate image enhancement and data interpretation. Despite numerous advancements in the field, there is a noticeable lack of studies that holistically address the entire microscopy acquisition chain. This review aims to highlight the potential of IoT and artificial intelligence (AI) in combination with portable microscopy, emphasizing the importance of a comprehensive approach to the microscopy acquisition chain, from portability to image analysis.

Fully Integrated and High-Throughput Microfluidic System for Multiplexed Point-Of-Care Testing

For every epidemic outbreak, the prevention and treatments in resource-limited areas are always out of reach. Critical to this is that high accuracy, stability, and more comprehensive analytical techniques always rely on expensive and bulky instruments and large laboratories. Here, a fully integrated and high-throughput microfluidic system is proposed for ultra-multiple point-of-care immunoassay, termed Dac system. Specifically, the Dac system only requires a handheld portable device to automatically recycle repetitive multi-step reactions including on-demand liquid releasing, dispensing, metering, collecting, oscillatory mixing, and discharging. The Dac system performs high-precision enzyme-linked immunosorbent assays for up to 17 samples or targets simultaneously on a single chip. Furthermore, reagent consumption is only 2% compared to conventional ELISA, and microbubble-accelerated reactions shorten the assay time by more than half. As a proof of concept, the multiplexed detections are achieved by detecting at least four infection targets for two samples simultaneously on a singular chip. Furthermore, the barcode-based multi-target results can rapidly distinguish between five similar cases, allowing for accurate therapeutic interventions. Compared to bulky clinical instruments, the accuracy of clinical inflammation classification is 92.38% (n = 105), with a quantitative correlation coefficient of R2 = 0.9838, while the clinical specificity is 100% and the sensitivity is 98.93%.

Artificial intelligence-powered electrochemical sensor: Recent advances, challenges, and prospects

Integrating artificial intelligence (AI) with electrochemical biosensors is revolutionizing medical treatments by enhancing patient data collection and enabling the development of advanced wearable sensors for health, fitness, and environmental monitoring. Electrochemical biosensors, which detect biomarkers through electrochemical processes, are significantly more effective. The integration of artificial intelligence is adept at identifying, categorizing, characterizing, and projecting intricate data patterns. As the Internet of Things (IoT), big data, and big health technologies move from theory to practice, AI-powered biosensors offer significant opportunities for real-time disease detection and personalized healthcare. Still, they also pose challenges such as data privacy, sensor stability, and algorithmic bias. This paper highlights the critical advances in material innovation, biorecognition elements, signal transduction, data processing, and intelligent decision systems necessary for developing next-generation wearable and implantable devices. Despite existing limitations, the integration of AI into biosensor systems shows immense promise for creating future medical devices that can provide early detection and improved patient outcomes, marking a transformative step forward in healthcare technology.

Effective Laboratory Diagnosis of Parasitic Infections of the Gastrointestinal Tract: Where, When, How, and What Should We Look For?

(1) Introduction: Gastrointestinal parasites (GIPs) are one of the most common causes of disease in the world. Clinical diagnosis of most parasitic diseases is difficult because they do not produce characteristic symptoms. (2) Methods: The PubMed, Science Direct, and Wiley Online Library medical databases were reviewed using the following phrases: "parasitic infections and diagnostics", "intestinal parasites", "gastrointestinal parasites", "parasitic infections and diagnostics", and their combinations. (3) Results and Conclusions: Correct diagnosis of GIP involves determining the presence of a parasite and establishing a relationship between parasite invasion and disease symptoms. The diagnostic process should consider the possibility of the coexistence of infection with several parasites at the same time. In such a situation, diagnostics should be planned with consideration of their frequency in each population and the local epidemiological situation. The importance of the proper interpretation of laboratory test results, based on good knowledge of the biology of the parasite, should be emphasized. The presence of the parasite may not be causally related to the disease symptoms. Due to wide access to laboratories, patients often decide to perform tests themselves without clinical justification. Research is carried out using various methods which are often unreliable. This review briefly covers current laboratory methods for diagnosing the most common gastrointestinal parasitic diseases in Europe. In particular, we provide useful information on the following aspects: (i) what to look for and where to look for it (suitability of feces, blood, duodenal contents, material taken from endoscopy or biopsy, tissue samples, and locations for searching for eggs, cysts, parasites, parasite genetic material, and characteristics of immune responses indicating parasitic infections); (ii) when material should be collected for diagnosis and/or to check the effectiveness of treatment; (iii) how-that is, by what methods-laboratory diagnostics should be carried out. Here, the advantages and disadvantages of direct and indirect methods of detecting parasites will be discussed. False-positive or false-negative results are a problem facing many tests. Available tests have different sensitivities and specificities. Therefore, especially in doubtful situations, tests for the presence of the pathogen should be performed using various available methods. It is important that the methods used make it possible to distinguish an active infection from a past infection. Finally, we present laboratory "case reports", in which we will discuss the diagnostic procedure that allows for the successful identification of parasites. Additionally, we briefly present the possibilities of using artificial intelligence to improve the effectiveness of diagnosing parasitic diseases.

Role of Machine Learning Assisted Biosensors in Point-of-Care-Testing For Clinical Decisions

Point-of-Care-Testing (PoCT) has emerged as an essential component of modern healthcare, providing rapid, low-cost, and simple diagnostic options. The integration of Machine Learning (ML) into biosensors has ushered in a new era of innovation in the field of PoCT. This article investigates the numerous uses and transformational possibilities of ML in improving biosensors for PoCT. ML algorithms, which are capable of processing and interpreting complicated biological data, have transformed the accuracy, sensitivity, and speed of diagnostic procedures in a variety of healthcare contexts. This review explores the multifaceted applications of ML models, including classification and regression, displaying how they contribute to improving the diagnostic capabilities of biosensors. The roles of ML-assisted electrochemical sensors, lab-on-a-chip sensors, electrochemiluminescence/chemiluminescence sensors, colorimetric sensors, and wearable sensors in diagnosis are explained in detail. Given the increasingly important role of ML in biosensors for PoCT, this study serves as a valuable reference for researchers, clinicians, and policymakers interested in understanding the emerging landscape of ML in point-of-care diagnostics.

A Biomimetic Optical Cardiac Fibrosis-on-a-Chip for High-Throughput Anti-Fibrotic Drug Screening

Cardiac fibrosis has emerged as the primary cause of morbidity, disability, and even mortality in numerous nations. In light of the advancements in precision medicine strategies, substantial attention has been directed toward the development of a practical and precise drug screening platform customized for individual patients. In this study, we introduce a biomimetic cardiac fibrosis-on-a-chip incorporating structural color hydrogels (SCHs) to enable optical high-throughput drug screening. By cocultivating a substantial proportion of cardiac fibroblasts (CFBs) with cardiomyocytes on the SCH, this biomimetic fibrotic microtissue successfully replicates the structural components and biomechanical properties associated with cardiac fibrosis. More importantly, the structural color shift observed in the SCH can be indicative of cardiac contraction and relaxation, making it a valuable tool for evaluating fibrosis progression. By incorporating such fibrotic microtissue into a microfluidic gradient chip, we develop a biomimetic optical cardiac fibrosis-on-a-chip platform that accurately and efficiently screens potential anti-fibrotic drugs. These characteristics suggest that this microphysiological platform possesses the capability to establish a preclinical framework for screening cardiac drugs, and may even contribute to the advancement of precision medicine.

High-Throughput and Integrated CRISPR/Cas12a-Based Molecular Diagnosis Using a Deep Learning Enabled Microfluidic System

CRISPR/Cas-based molecular diagnosis demonstrates potent potential for sensitive and rapid pathogen detection, notably in SARS-CoV-2 diagnosis and mutation tracking. Yet, a major hurdle hindering widespread practical use is its restricted throughput, limited integration, and complex reagent preparation. Here, a system, microfluidic multiplate-based ultrahigh throughput analysis of SARS-CoV-2 variants of concern using CRISPR/Cas12a and nonextraction RT-LAMP (mutaSCAN), is proposed for rapid detection of SARS-CoV-2 and its variants with limited resource requirements. With the aid of the self-developed reagents and deep-learning enabled prototype device, our mutaSCAN system can detect SARS-CoV-2 in mock swab samples below 30 min as low as 250 copies/mL with the throughput up to 96 per round. Clinical specimens were tested with this system, the accuracy for routine and mutation testing (22 wildtype samples, 26 mutational samples) was 98% and 100%, respectively. No false-positive results were found for negative (n = 24) samples.

AI-powered microscopy image analysis for parasitology: integrating human expertise

Microscopy image analysis plays a pivotal role in parasitology research. Deep learning (DL), a subset of artificial intelligence (AI), has garnered significant attention. However, traditional DL-based methods for general purposes are data-driven, often lacking explainability due to their black-box nature and sparse instructional resources. To address these challenges, this article presents a comprehensive review of recent advancements in knowledge-integrated DL models tailored for microscopy image analysis in parasitology. The massive amounts of human expert knowledge from parasitologists can enhance the accuracy and explainability of AI-driven decisions. It is expected that the adoption of knowledge-integrated DL models will open up a wide range of applications in the field of parasitology.

Self-reporting electroswitchable colorimetric platform for smart ammonium recovery from wastewater

Recovery of ammonium from wastewater represents a sustainable strategy within the context of global resource depletion, environmental pollution and carbon neutralization. The present study developed an advanced self-reporting electroswitchable colorimetric platform (SECP) to realize smart ammonium recovery based on the electrically stimulated transformation of Prussian blue/Prussian white (PB/PW) redox couple. The key to SECP was the selectivity of ammonium adsorption, sensitivity of desorption to electric signals and visualability of color change during switchable adsorption/desorption transformation. The results demonstrated the electrochemical intercalation-induced selective adsorption of NH4+ (selectivity coefficient of 3-19 versus other cations) and deintercalation-induced desorption on the PB-film electrode. At applied voltage of 1.2 V for 20 min, the negatively charged PB-film electrode achieved the maximum adsorption capacity of 3.2 mmol g-1. Reversing voltage to -0.2 V for 20 min resulted in desorption efficiency as high as 99%, indicating high adsorption/desorption reversibility and cyclic stability. The Fe(III)/Fe(II) redox dynamics were responsible for PB/PW transformation during reversible intercalation/deintercalation of NH4+. Based on the blue/transparence color change of PB/PW, the quantitative relationship was established between amounts of NH4+ adsorbed and extracted RGB values by multiple linear regression (R2 = 0.986, RMSE = 0.095). Then, the SECP was created upon the unique capability of real-time monitoring and feedback of color change of electrode to realize the automatic control of NH4+ adsorption/desorption. During five cycles of tests, the adsorption process consistently peaked at an average value of 3.15±0.04 mmol g-1, while desorption reliably approached the near-zero average of 0.06±0.04 mmol g-1. The average time of duration was 19.6±1.67 min for adsorption and 18.8±1.10 min for desorption, respectively. With electroswitchability, selectivity and self-reporting functionalities, the SECP represents a paradigm shift in smart ammonium recovery from wastewater, making wastewater treatment and resource recovery more efficient, more intelligent and more sustainable.

A smartphone serves as a data logger for a fully automated lab-constructed microfluidic system

Fluorescence is an innovative technique that has captivated scholars in recent years due to its superior sensitivity and selectivity. The development of microfluidic components has added to its appeal, particularly given the technology ability to control fluid using very small quantities (microliter range) and achieve high liquid throughput. We have combined these two technologies to develop a lab-constructed simple system for measuring fluorescence, notable for the following features:•The device constructed entirely in our lab and programmed for measuring the fluorescence of liquids using microfluidic technology, delivered excellent results. The regression coefficient R² (0.9995) was obtained five points between 0.001-0.01µg .ml-1. Moreover, the reproducibility standard deviation (%) of 0.008 µg .ml-1 fluorescein dye remained at zero, for ten repeated experiments.•The device was full automated using a smartphone as a data logger, and lab-constructed programs.•The results were satisfactory with a detection limit of 1 × 10-4 µg.ml-1. This proposed system can measure over 200 samples per hour making it highly efficient and eco-friendly due to the reduced use of reagents and lower waste production. The fully automated system can effectively be used to determine fluorescein dye concentrations. Another application (micro pump view) manages all actions required in this microfluidic system, such as operating the two lab-constructed peristaltic pumps.

A Highly Sensitive Dual-Drive Microfluidic Device for Multiplexed Detection of Respiratory Virus Antigens

Conventional microfluidic systems that rely on capillary force have a fixed structure and limited sensitivity, which cannot meet the demands of clinical applications. Herein, we propose a dual-drive microfluidic device for sensitive and flexible detection of multiple pathogenic microorganisms antigens/antibodies. The device comprises a portable microfluidic analyzer and a dual-drive microfluidic chip. Along with capillary force, a second active driving force is provided by a removable self-driving valve in the waste chamber. The interval between these two driving forces can be adjusted to control the reaction time in the microchannel, optimizing the formation of antigen-antibody complexes and enhancing sensitivity. Moreover, the material used in the self-driving valve can be changed to adjust the active force strength needed for different tests. The device offers quantitative analysis for respiratory syncytial virus antigen and SARS-CoV-2 antigen using a 35 μL sample, delivering results within 5 min. The detection limits of the system were 1.121 ng/mL and 0.447 ng/mL for respiratory syncytial virus recombinant fusion protein and SARS-CoV-2 recombinant nucleoprotein, respectively. Although the dual-drive microfluidic device has been used for immunoassay for respiratory syncytial virus and SARS-CoV-2 in this study, it can be easily adapted to other immunoassay applications by changing the critical reagents.

A smartphone-based diagnostic analyzer for point-of-care milk somatic cell counting

BACKGROUND

Mastitis, a pervasive and detrimental disease in dairy farming, poses a significant challenge to the global dairy industry. Monitoring the milk somatic cell count (SCC) is vital for assessing the incidence of mastitis and the quality of raw cow's milk. However, existing SCC detection methods typically require large-scale instruments and specialized operators, limiting their application in resource-constrained settings such as dairy farms and small-scale labs. To address these limitations, this study introduces a novel, smartphone-based, on-site SCC testing method that leverages smartphone capabilities for milk somatic cell identification and enumeration, offering a portable and user-friendly testing platform.

RESULTS

The central findings of our study demonstrate the effectiveness of the proposed method for counting milk somatic cells. Its on-site applicability, facilitated by the microfluidic chip, optical system, and smartphone integration, heralds a paradigm shift in point-of-care testing (POCT) for dairy farms and smaller laboratories. This approach bypasses complex processing and presents a user-friendly solution for real-time SCC monitoring in resource-limited settings. This device boasts several unique features: small size, low cost (<$1,000 total manufacturing cost and <$1 per test), and high accuracy. Remarkably, it delivers test results within just 2 min. Actual-sample testing confirmed its consistency with results from the commercial Bentley FTS/FCM cytometer, affirming the reliability of the proposed method. Overall, these results underscore the potential for transformative change in dairy farm management and laboratory testing practices.

SIGNIFICANCE

In summary, this study concludes that the proposed smartphone-based method significantly contributes to the accessibility and ease of SCC testing in resource-limited environments. By fostering the use of POCT technology in food safety control, particularly in the dairy industry, this innovative approach has the potential to revolutionize the monitoring and management of mastitis, ultimately benefiting the global dairy sector.

Advances in Nucleic Acid Assays for Infectious Disease: The Role of Microfluidic Technology

Within the fields of infectious disease diagnostics, microfluidic-based integrated technology systems have become a vital technology in enhancing the rapidity, accuracy, and portability of pathogen detection. These systems synergize microfluidic techniques with advanced molecular biology methods, including reverse transcription polymerase chain reaction (RT-PCR), loop-mediated isothermal amplification (LAMP), and clustered regularly interspaced short palindromic repeats (CRISPR), have been successfully used to identify a diverse array of pathogens, including COVID-19, Ebola, Zika, and dengue fever. This review outlines the advances in pathogen detection, attributing them to the integration of microfluidic technology with traditional molecular biology methods and smartphone- and paper-based diagnostic assays. The cutting-edge diagnostic technologies are of critical importance for disease prevention and epidemic surveillance. Looking ahead, research is expected to focus on increasing detection sensitivity, streamlining testing processes, reducing costs, and enhancing the capability for remote data sharing. These improvements aim to achieve broader coverage and quicker response mechanisms, thereby constructing a more robust defense for global public health security.

Nucleic Acid Plate Culture: Label-Free and Naked-Eye-Based Digital Loop-Mediated Isothermal Amplification in Hydrogel with Machine Learning

Digital nucleic acid amplification enables the absolute quantification of single molecules. However, due to the ultrasmall reaction volume in the digital system (i.e., short light path), most digital systems are limited to fluorescence signals, while label-free and naked-eye readout remain challenging. In this work, we report a digital nucleic acid plate culture method for label-free, ultrasimple, and naked-eye nucleic acid analysis. As simple as the bacteria culture, the nanoconfined digital loop-mediated isothermal amplification was performed by using polyacrylamide (PAM) hydrogel as the amplification matrix. The nanoconfinement of PAM hydrogel with an ionic polymer chain can remarkably accelerate the amplification of target nucleic acids and the growth of inorganic byproducts, namely, magnesium pyrophosphate particles (MPPs). Compared to that in aqueous solutions, MPPs trapped in the hydrogel with enhanced light scattering characteristics are clearly visible to the naked eye, forming white "colony" spots that can be simply counted in a label-free and instrument-free manner. The MPPs can also be photographed by a smartphone and automatically counted by a machine-learning algorithm to realize the absolute quantification of antibiotic-resistant pathogens in diverse real samples.

Quantitative, high-sensitivity measurement of liquid analytes using a smartphone compass

Smartphone ubiquity has led to rapid developments in portable diagnostics. While successful, such platforms are predominantly optics-based, using the smartphone camera as the sensing interface. By contrast, magnetics-based modalities exploiting the smartphone compass (magnetometer) remain unexplored, despite inherent advantages in optically opaque, scattering or auto-fluorescing samples. Here we report smartphone analyte sensing utilizing the built-in magnetometer for signal transduction via analyte-responsive magnetic-hydrogel composites. As these hydrogels dilate in response to targeted stimuli, they displace attached magnetic material relative to the phone's magnetometer. Using a bilayer hydrogel geometry to amplify this motion allows for sensitive, optics-free, quantitative liquid-based analyte measurements that require neither any electronics nor power beyond that contained within the smartphone itself. We demonstrate this concept with glucose-specific and pH-responsive hydrogels, including glucose detection down to single-digit micromolar concentrations with potential for extension to nanomolar sensitivities. The platform is adaptable to numerous measurands, opening a path towards portable, inexpensive sensing of multiple analytes or biomarkers of interest.

Smartphone-based device for point-of-care diagnostics of pulmonary inflammation using convolutional neural networks (CNNs)

In pulmonary inflammation diseases, like COVID-19, lung involvement and inflammation determine the treatment regime. Respiratory inflammation is typically arisen due to the cytokine storm and the leakage of the vessels for immune cells recruitment. Currently, such a situation is detected by the clinical judgment of a specialist or precisely by a chest CT scan. However, the lack of accessibility to the CT machines in many poor medical centers as well as its expensive service, demands more accessible methods for fast and cheap detection of lung inflammation. Here, we have introduced a novel method for tracing the inflammation and lung involvement in patients with pulmonary inflammation, such as COVID-19, by a simple electrolyte detection in their sputum samples. The presence of the electrolyte in the sputum sample results in the fern-like structures after air-drying. These fern patterns are different in the CT positive and negative cases that are detected by an AI application on a smartphone and using a low-cost and portable mini-microscope. Evaluating 160 patient-derived sputum sample images, this method demonstrated an interesting accuracy of 95%, as confirmed by CT-scan results. This finding suggests that the method has the potential to serve as a promising and reliable approach for recognizing lung inflammatory diseases, such as COVID-19.

Convolutional Neural Network-Driven Impedance Flow Cytometry for Accurate Bacterial Differentiation

Impedance flow cytometry (IFC) has been demonstrated to be an efficient tool for label-free bacterial investigation to obtain the electrical properties in real time. However, the accurate differentiation of different species of bacteria by IFC technology remains a challenge owing to the insignificant differences in data. Here, we developed a convolutional neural networks (ConvNet) deep learning approach to enhance the accuracy and efficiency of the IFC toward distinguishing various species of bacteria. First, more than 1 million sets of impedance data (comprising 42 characteristic features for each set) of various groups of bacteria were trained by the ConvNet model. To improve the efficiency for data analysis, the Spearman correlation coefficient and the mean decrease accuracy of the random forest algorithm were introduced to eliminate feature interaction and extract the opacity of impedance related to the bacterial wall and membrane structure as the predominant features in bacterial differentiation. Moreover, the 25 optimized features were selected with differentiation accuracies of >96% for three groups of bacteria (bacilli, cocci, and vibrio) and >95% for two species of bacilli (Escherichia coli and Salmonella enteritidis), compared to machine learning algorithms (complex tree, linear discriminant, and K-nearest neighbor algorithms) with a maximum accuracy of 76.4%. Furthermore, bacterial differentiation was achieved on spiked samples of different species with different mixing ratios. The proposed ConvNet deep learning-assisted data analysis method of IFC exhibits advantages in analyzing a huge number of data sets with capacity for extracting predominant features within multicomponent information and will bring about progress and advances in the fields of both biosensing and data analysis.

Virus detection light diffraction fingerprints for biological applications

The transmission of viral diseases is highly unstable and highly contagious. As the carrier of virus transmission, cell is an important factor to explore the mechanism of virus transmission and disease. However, there is still a lack of effective means to continuously monitor the process of viral infection in cells, and there is no rapid, high-throughput method to assess the status of viral infection. On the basis of the virus light diffraction fingerprint of cells, we applied the gray co-occurrence matrix, set the two parameters effectively to distinguish the virus status and infection time of cells, and visualized the virus infection process of cells in high throughput. We provide an efficient and nondestructive testing method for the selection of excellent livestock and poultry breeds at the cellular level. Meanwhile, our work provides detection methods for the recessive transmission of human-to-human, animal-to-animal, and zoonotic diseases and to inhibit and block their further development.

Detection of Parasites in the Field: The Ever-Innovating CRISPR/Cas12a

The rapid and accurate identification of parasites is crucial for prompt therapeutic intervention in parasitosis and effective epidemiological surveillance. For accurate and effective clinical diagnosis, it is imperative to develop a nucleic-acid-based diagnostic tool that combines the sensitivity and specificity of nucleic acid amplification tests (NAATs) with the speed, cost-effectiveness, and convenience of isothermal amplification methods. A new nucleic acid detection method, utilizing the clustered regularly interspaced short palindromic repeats (CRISPR)-associated (Cas) nuclease, holds promise in point-of-care testing (POCT). CRISPR/Cas12a is presently employed for the detection of Plasmodium falciparum, Toxoplasma gondii, Schistosoma haematobium, and other parasites in blood, urine, or feces. Compared to traditional assays, the CRISPR assay has demonstrated notable advantages, including comparable sensitivity and specificity, simple observation of reaction results, easy and stable transportation conditions, and low equipment dependence. However, a common issue arises as both amplification and cis-cleavage compete in one-pot assays, leading to an extended reaction time. The use of suboptimal crRNA, light-activated crRNA, and spatial separation can potentially weaken or entirely eliminate the competition between amplification and cis-cleavage. This could lead to enhanced sensitivity and reduced reaction times in one-pot assays. Nevertheless, higher costs and complex pre-test genome extraction have hindered the popularization of CRISPR/Cas12a in POCT.

A smartphone-based centrifugal mHealth platform implementing hollow daisy-shaped quick response chip for hematocrit measurement

Due to the ever-increasing challenge of emerging and reemerging infections on global health, the development of POCT tools has been propelled. However, conventional point-of-care testing methods suffer from several limitations, including cumbersome operation, long detection times, and low accuracy, which hamper their widespread application. Compared to traditional disease diagnostic equipment, mobile health platforms offer several advantages, including portability, ease of operation, and automated analysis of detection results through recognition algorithms. Consequently, they hold great promise for the future. Here, we developed a smartphone-based centrifugal mHealth platform implementing daisy-shaped quick response chip for hematocrit measurement. The centrifugal microfluidic chip is combined with a smartphone through a back-clip-on mobile phone adapter whose control circuit is designed with low power consumption to enable the platform to operate without requiring a high-power source that is inconvenient to carry, thereby achieving the goal of portability. Concurrently, we designed a quick response chip featuring a unique hollow daisy structure that is in line with the properties of hematocrit detection. The distinctive configuration of the chip enables adequate centrifugal force to be supplied for hematocrit detection. Additionally, our customized quick response code recognition algorithm is able to recognize this chip, facilitating non-experts in performing hematocrit intelligent recognition with their smartphones.

POC device for rapid oral pH determination based on a smartphone platform

Salivary pH serves as a valuable and useful diagnostic marker for periodontal disease, as it not only plays a critical role in disease prevention but also in its development. Typically, saliva sampling is collected by draining and spitting it into collection tubes or using swabs. In this study, we have developed a Point-of-Care (POC) device for in situ determination of oral pH without the need for complex instruments, relying solely on a smartphone as the detection device. Our system utilizes a non-toxic vegetable colourimetric indicator, immobilized on a chitosan membrane located on a disposable stick, enabling direct sampling within the buccal cavity. An ad hoc designed 3D-printed attachment is used to ensure accurate positioning and alignment of the stick, as well as isolation from external lighting conditions. A custom-developed smartphone application captures and automatically processes the image of the sensing membrane, providing the salivary pH results. After optimizing the cocktail composition, the developed sensors demonstrated the capacity to determine pH within a range of 5.4 to 8.1 with a remarkable precision of 0.6%, achieving a very short analysis time of just 1 min. A stability study conducted on the sensing membranes revealed a lifetime of 50 days. To validate the performance of our analytical device, we compared its results against those obtained from a calibrated pH-meter, using a group of individuals. The device exhibited an average error of 2.4% when compared with the pH-meter results, confirming its reliability and accuracy.

Minimum Minutes Machine-Learning Microfluidic Microbe Monitoring Method (M7)

Frequent outbreaks of viral diseases have brought substantial negative impacts on society and the economy, and they are very difficult to detect, as the concentration of viral aerosols in the air is low and the composition is complex. The traditional detection method is manually collection and re-detection, being cumbersome and time-consuming. Here we propose a virus aerosol detection method based on microfluidic inertial separation and spectroscopic analysis technology to rapidly and accurately detect aerosol particles in the air. The microfluidic chip is designed based on the principles of inertial separation and laminar flow characteristics, resulting in an average separation efficiency of 95.99% for 2 μm particles. We build a microfluidic chip composite spectrometer detection platform to capture the spectral information on aerosol particles dynamically. By employing machine-learning techniques, we can accurately classify different types of aerosol particles. The entire experiment took less than 30 min as compared with hours by PCR detection. Furthermore, our model achieves an accuracy of 97.87% in identifying virus aerosols, which is comparable to the results obtained from PCR detection.

Microsnoop: A generalist tool for microscopy image representation

Accurate profiling of microscopy images from small scale to high throughput is an essential procedure in basic and applied biological research. Here, we present Microsnoop, a novel deep learning-based representation tool trained on large-scale microscopy images using masked self-supervised learning. Microsnoop can process various complex and heterogeneous images, and we classified images into three categories: single-cell, full-field, and batch-experiment images. Our benchmark study on 10 high-quality evaluation datasets, containing over 2,230,000 images, demonstrated Microsnoop's robust and state-of-the-art microscopy image representation ability, surpassing existing generalist and even several custom algorithms. Microsnoop can be integrated with other pipelines to perform tasks such as superresolution histopathology image and multimodal analysis. Furthermore, Microsnoop can be adapted to various hardware and can be easily deployed on local or cloud computing platforms. We will regularly retrain and reevaluate the model using community-contributed data to consistently improve Microsnoop.

Recent advances in point-of-care testing of COVID-19

Advances in microfluidic device miniaturization and system integration contribute to the development of portable, handheld, and smartphone-compatible devices. These advancements in diagnostics have the potential to revolutionize the approach to detect and respond to future pandemics. Accordingly, herein, recent advances in point-of-care testing (POCT) of coronavirus disease 2019 (COVID-19) using various microdevices, including lateral flow assay strips, vertical flow assay strips, microfluidic channels, and paper-based microfluidic devices, are reviewed. However, visual determination of the diagnostic results using only microdevices leads to many false-negative results due to the limited detection sensitivities of these devices. Several POCT systems comprising microdevices integrated with portable optical readers have been developed to address this issue. Since the outbreak of COVID-19, effective POCT strategies for COVID-19 based on optical detection methods have been established. They can be categorized into fluorescence, surface-enhanced Raman scattering, surface plasmon resonance spectroscopy, and wearable sensing. We introduced next-generation pandemic sensing methods incorporating artificial intelligence that can be used to meet global health needs in the future. Additionally, we have discussed appropriate responses of various testing devices to emerging infectious diseases and prospective preventive measures for the post-pandemic era. We believe that this review will be helpful for preparing for future infectious disease outbreaks.

Next-Generation Microfluidics for Biomedical Research and Healthcare Applications

Microfluidic systems offer versatile biomedical tools and methods to enhance human convenience and health. Advances in these systems enables next-generation microfluidics that integrates automation, manipulation, and smart readout systems, as well as design and three-dimensional (3D) printing for precise production of microchannels and other microstructures rapidly and with great flexibility. These 3D-printed microfluidic platforms not only control the complex fluid behavior for various biomedical applications, but also serve as microconduits for building 3D tissue constructs-an integral component of advanced drug development, toxicity assessment, and accurate disease modeling. Furthermore, the integration of other emerging technologies, such as advanced microscopy and robotics, enables the spatiotemporal manipulation and high-throughput screening of cell physiology within precisely controlled microenvironments. Notably, the portability and high precision automation capabilities in these integrated systems facilitate rapid experimentation and data acquisition to help deepen our understanding of complex biological systems and their behaviors. While certain challenges, including material compatibility, scaling, and standardization still exist, the integration with artificial intelligence, the Internet of Things, smart materials, and miniaturization holds tremendous promise in reshaping traditional microfluidic approaches. This transformative potential, when integrated with advanced technologies, has the potential to revolutionize biomedical research and healthcare applications, ultimately benefiting human health. This review highlights the advances in the field and emphasizes the critical role of the next generation microfluidic systems in advancing biomedical research, point-of-care diagnostics, and healthcare systems.

Deep-Learning Terahertz Single-Cell Metabolic Viability Study

Cell viability assessment is critical, yet existing assessments are not accurate enough. We report a cell viability evaluation method based on the metabolic ability of a single cell. Without culture medium, we measured the absorption of cells to terahertz laser beams, which could target a single cell. The cell viability was assessed with a convolution neural classification network based on cell morphology. We established a cell viability assessment model based on the THz-AS (terahertz-absorption spectrum) results as y = a = (x - b)c, where x is the terahertz absorbance and y is the cell viability, and a, b, and c are the fitting parameters of the model. Under water stress the changes in terahertz absorbance of cells corresponded one-to-one with the apoptosis process, and we propose a cell 0 viability definition as terahertz absorbance remains unchanged based on the cell metabolic mechanism. Compared with typical methods, our method is accurate, label-free, contact-free, and almost interference-free and could help visualize the cell apoptosis process for broad applications including drug screening.

Plant exosome nanovesicles (PENs): green delivery platforms

Natural plants have been attracting increasing attention in biomedical research due to their numerous benefits. Plant exosome-derived vesicles, some of the plant's components, are small nanoscale vesicles secreted by plant cells. These vesicles are rich in bioactive substances and play significant roles in intercellular communication, information transfer, and maintaining homeostasis in organisms. They also hold promise for treating diseases, and their vesicular structures make them suitable carriers for drug delivery, with large-scale production feasible. Therefore, this paper aims to provide an overview of nanovesicles from different plant sources and their extraction methods. We also outline the biological activities of nanovesicles, including their anti-inflammatory, anti-viral, and anti-tumor properties, and systematically introduce their applications in drug delivery. These applications include transdermal delivery, targeted drug delivery, gene delivery, and their potential use in the modern food industry. This review provides new ideas and methods for future research on plant exosomes, including their empowerment by artificial intelligence and gene editing, as well as their potential application in the biomedicine, food, and agriculture industries.

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.

Enhancing single-cell biology through advanced AI-powered microfluidics

Microfluidic technology has largely benefited both fundamental biological research and translational clinical diagnosis with its advantages in high-throughput, single-cell resolution, high integrity, and wide-accessibility. Despite the merits we obtained from microfluidics in the last two decades, the current requirement of intelligence in biomedicine urges the microfluidic technology to process biological big data more efficiently and intelligently. Thus, the current readout technology based on the direct detection of the signals in either optics or electrics was not able to meet the requirement. The implementation of artificial intelligence (AI) in microfluidic technology matches up with the large-scale data usually obtained in the high-throughput assays of microfluidics. At the same time, AI is able to process the multimodal datasets obtained from versatile microfluidic devices, including images, videos, electric signals, and sequences. Moreover, AI provides the microfluidic technology with the capability to understand and decipher the obtained datasets rather than simply obtaining, which eventually facilitates fundamental and translational research in many areas, including cell type discovery, cell signaling, single-cell genetics, and diagnosis. In this Perspective, we will highlight the recent advances in employing AI for single-cell biology and present an outlook on the future direction with more advanced AI algorithms.

Smartphone-Based High-Throughput Fluorimetric Assay for Histidine Quantification in Human Urine Using 96-Well Plates

A high-throughput fluorimetric assay for histidine was developed, using a 96-well plates platform. The analyte reacts selectively with o-phthalaldehyde under mild alkaline conditions to form a stable derivative. Instrumental-free detection was carried out using a smartphone after illumination under UV light (365 nm). The method was proved to be linear up to 100 μM histidine, with an LLOQ (lower limit of quantification) of 10 μM. The assay was only prone to interference from glutathione and histamine that exist in the urine samples at levels that are orders of magnitude lower compared to histidine. Human urine samples were analyzed following minimum treatment and were found to contain histidine in the range of 280 to 1540 μM. The results were in good agreement with an HPLC corroborative method.

Revolutionizing microfluidics with artificial intelligence: a new dawn for lab-on-a-chip technologies

Keisuke Goda, Hang Lu, Peng Fei, and Jochen Guck introduce the AI in Microfluidics themed collection, on revolutionizing microfluidics with artificial intelligence: a new dawn for lab-on-a-chip technologies.

Engineering Innovative Interfaces for Point-of-Care Diagnostics

The ongoing Coronavirus disease 2019 (COVID-19) pandemic illustrates the need for sensitive and reliable tools to diagnose and monitor diseases. Traditional diagnostic approaches rely on centralized laboratory tests that result in long wait times to results and reduce the number of tests that can be given. Point-of-care tests (POCTs) are a group of technologies that miniaturize clinical assays into portable form factors that can be run both in clinical areas --in place of traditional tests-- and outside of traditional clinical settings --to enable new testing paradigms. Hallmark examples of POCTs are the pregnancy test lateral flow assay and the blood glucose meter. Other uses for POCTs include diagnostic assays for diseases like COVID-19, HIV, and malaria but despite some successes, there are still unsolved challenges for fully translating these lower cost and more versatile solutions. To overcome these challenges, researchers have exploited innovations in colloid and interface science to develop various designs of POCTs for clinical applications. Herein, we provide a review of recent advancements in lateral flow assays, other paper based POCTs, protein microarray assays, microbead flow assays, and nucleic acid amplification assays. Features that are desirable to integrate into future POCTs, including simplified sample collection, end-to-end connectivity, and machine learning, are also discussed in this review.

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.

Handyfuge Microfluidic for On-Site Antibiotic Susceptibility Testing

Low-cost, rapid, and accurate acquisition of minimum inhibitory concentrations (MICs) is key to limiting the development of antimicrobial resistance (AMR). Until now, conventional antibiotic susceptibility testing (AST) methods are typically time-consuming, high-cost, and labor-intensive, making them difficult to accomplish this task. Herein, an electricity-free, portable, and robust handyfuge microfluidic chip was developed for on-site AST, termed handyfuge-AST. With simply handheld centrifugation, the bacterial-antibiotic mixtures with accurate antibiotic concentration gradients could be generated in less than 5 min. The accurate MIC values of single antibiotics (including ampicillin, kanamycin, and chloramphenicol) or their combinations against Escherichia coli could be obtained within 5 h. To further meet the growing demands of point-of-care testing, we upgraded our handyfuge-AST with a pH-based colorimetric strategy, enabling naked eye recognition or intelligent recognition with a homemade mobile app. Through a comparative study of 60 clinical data (10 clinical samples corresponding to six commonly used antibiotics), the accurate MICs by handyfuge-AST with 100% categorical agreements were achieved compared to clinical standard methods (area under curves, AUCs = 1.00). The handyfuge-AST could be used as a low-cost, portable, and robust point-of-care device to rapidly obtain accurate MIC values, which significantly limit the progress of AMR.