Rapid deep learning-assisted predictive diagnostics for point-of-care testing

Artificial intelligence-enabled microsphere imaging immunosensor based on magnetic metal-organic frameworks-assisted sample pretreatment for detecting aflatoxin B1 in peanuts

Sensitive and rapid detection of aflatoxin B1 (AFB1) is vital for safeguarding food safety, considering its potent carcinogenic toxicity. Herein, an artificial intelligence-enabled microsphere imaging (AI-MI) immunosensor based on magnetic metal-organic frameworks-assisted sample pretreatment was developed for detecting AFB1 in peanuts. In this work, Fe3O4@MIL-101(Fe) served as a magnetic adsorbent to efficiently enrich AFB1. Based on the competitive immunoreaction, the enriched AFB1 modulated the amount of horseradish peroxidase (HRP)-labeled goat anti-mouse antibody conjugated on the polystyrene (PS) immuno-microsphere. The HRP can catalyze the rapid formation of polydopamine on the surface of the PS microsphere with additional hydrogen peroxide. Due to the abundant functional groups, the polydopamine coating could adsorb amino-functionalized magnetic nanoparticles to form PS probes. The PS probes were magnetically separated, visualized with an optical microscope, and counted using a computer vision algorithm. Finally, the changes in the number of PS probes were correlated with the amount of AFB1. Under optimized conditions, Fe3O4@MIL-101(Fe) exhibited remarkable enrichment capacity (1.59 mg/g), and the AI-MI immunosensor showed a high sensitivity (4.90 pg/mL, 19-fold improvement over enzyme-linked immunosorbent assay) and a wide linear range (from 0.01 to 500 ng/mL) for AFB1. This AI-MI immunosensor holds significant promise for intelligent detection of trace toxins.

Artificial intelligence in bacterial diagnostics and antimicrobial susceptibility testing: Current advances and future prospects

Recently, artificial intelligence (AI) has emerged as a transformative tool, enhancing the speed, accuracy, and scalability of bacterial diagnostics. This review explores the role of AI in revolutionizing bacterial detection and antimicrobial susceptibility testing (AST) by leveraging machine learning models, including Random Forest, Support Vector Machines (SVM), and deep learning architectures such as Convolutional Neural Networks (CNNs) and transformers. The integration of AI into these methods promises to address the current limitations of traditional techniques, offering a path toward more efficient, accessible, and reliable diagnostic solutions. In particular, AI-based approaches have demonstrated significant potential in resource-limited settings by enabling cost-effective and portable diagnostic solutions, reducing dependency on specialized infrastructure, and facilitating remote bacterial detection through smartphone-integrated platforms and telemedicine applications. This review highlights AI's transformative role in automating data analysis, minimizing human error, and delivering real-time diagnostic results, ultimately improving patient outcomes and optimizing healthcare efficiency. In addition, we not only examine the current advances in machine learning and deep learning but also review their applications in plate counting, mass spectrometry, morphology-based and motion-based microscopic detection, holographic microscopy, colorimetric and fluorescence detection, electrochemical sensors, Raman and Surface-Enhanced Raman Spectroscopy (SERS), and Atomic Force Microscopy (AFM) for bacterial diagnostics and AST. Finally, we discuss the future directions and potential advancements in AI-driven bacterial diagnostics.

Artificial intelligence (AI) in point-of-care testing

The integration of artificial intelligence (AI) into point-of-care testing (POCT) represents a transformative leap in modern healthcare, addressing critical challenges in diagnostic accuracy, workflow efficiency, and equitable access. While POCT has revolutionized decentralized care through rapid results, its potential is hindered by variability in accuracy, integration hurdles, and resource constraints. AI technologies-encompassing machine learning, deep learning, and natural language processing-offer robust solutions: convolutional neural networks improve malaria detection in sub-Saharan Africa to 95 % sensitivity, while predictive analytics reduce device downtime by 20 % in resource-limited settings. AI-driven decision support systems curtail antibiotic misuse by 40 % through real-time data synthesis, and portable AI devices enable anaemia screening in rural India with 94 % accuracy, slashing diagnostic delays from weeks to hours. Despite these advancements, challenges persist, including data privacy risks, algorithmic opacity, and infrastructural gaps in low- and middle-income countries. Explainable AI frameworks and blockchain encryption are critical to building clinician trust and ensuring regulatory compliance. Future directions emphasize the convergence of AI with Internet of Things (IoT) and blockchain for predictive diagnostics, as demonstrated by AI-IoT systems forecasting dengue outbreaks 14 days in advance. Personalized medicine, powered by genomic and wearable data integration, further underscores AI potential to tailor therapies, reducing cardiovascular events by 25 %. Realizing this vision demands interdisciplinary collaboration, ethical governance, and equitable implementation to bridge global health disparities. By harmonizing innovation with accessibility, AI-enhanced POCT emerges as a cornerstone of proactive, patient-centered healthcare, poised to democratize diagnostics and drive sustainable health equity worldwide.

Dynamic Prediction and Intervention of Serum Sodium in Patients with Stroke Based on Attention Mechanism Model

Abnormal serum sodium levels are a common and severe complication in stroke patients, significantly increasing mortality risk and prolonging ICU stays. Accurate real-time prediction of serum sodium fluctuations is crucial for optimizing clinical interventions. However, existing predictive models face limitations in handling complex dynamic features and long time series data, making them less effective in guiding individualized treatment. To address this challenge, this study developed a deep learning model based on a multi-head attention mechanism to enable real-time prediction of serum sodium concentrations and provide personalized intervention recommendations for ICU stroke patients. This study utilized publicly available MIMIC-III (n = 2346) and MIMIC-IV (n = 896) datasets, extracting time series data from 10 key clinical indicators closely associated with serum sodium levels. To address the complexity of long time series data, a moving sliding window sub-sampling segmentation method was employed, effectively transforming extensive sequences into more manageable inputs while preserving critical temporal dependencies. By leveraging advanced mathematical modeling, meaningful insights were extracted from sparse and irregular time series data. The resulting time-feature fusion multi-head attention (TFF-MHA) model underwent rigorous validation using public datasets and demonstrated superior performance in predicting both serum sodium values and corresponding intervention measures compared to existing models. This study contributes to the field of healthcare informatics by introducing an innovative, data-driven approach for dynamic serum sodium prediction and intervention recommendation, providing a valuable clinical decision-support tool for optimizing sodium management strategies in critically ill stroke patients.

Rapid antimicrobial susceptibility tests performed by self-diluting microfluidic chips for drug resistance studies and point-of-care diagnostics

Antimicrobial resistance (AMR) is a global public health issue. Rapid and accurate antimicrobial susceptibility tests (AST) on bacteria isolates would facilitate appropriate choice of antibiotics, in which patients receive appropriate treatment and the emergence of multidrug-resistant organisms could be prevented simultaneously. In this study, we have developed a microfluidic device named Self Dilution for Faster Antimicrobial Susceptibility Testing (SDFAST). This SlipChip-based device consists of two layers of microchips, allowing injection of bacterial suspension and antibiotics by simply connecting the two chips. By slipping one microchip against another in a single press of the microchip, the antibiotics can be diluted within seconds and be well mixed with bacterial samples. By combining SDFAST with a water-soluble tetrazolium salt-8 (WST-8) assay, a range of clinically prevalent bacteria, including Acinetobacter baumannii, Escherichia coli, Klebsiella pneumoniae, and Staphylococci species, were tested under various antibiotics. Color analysis after 4-6 h of incubation showed an abrupt change in the WST-8 color of certain wells with diluted antibiotics, proving that instrument-free and immediate identification of minimum inhibitory concentration (MIC) could be achieved. The testing on 51 clinical isolates had an agreement of 92%, proving the accuracy of our method. These results validated its advantages of simple operation, rapid testing, and low sample consumption comparing to conventional methods, which require 16-24 h of incubation. Therefore, our method shows great potential to be further developed into a medical instrument for automated medical testing and point-of-care diagnosis.

Point-of-Care Testing: The Convergence of Innovation and Accessibility in Diagnostics

Over the years, the evolution of point-of-care testing (POCT) has been driven by technological advancements in materials, design, and artificial intelligence, as well as breakthrough developments in wearable technologies. These innovations are shifting diagnostics from centralized medical facilities to individual homes, meeting the growing demand for personalized healthcare. This Review explores recent advancements in binding-based assay technologies over the past two years, focusing on platforms such as traditional flow assays (lateral and vertical flow), fully integrated microfluidic devices, and wearable biosensor-integrated systems for POCT applications. It emphasizes the role of optical and electrochemical detection methods, which are essential for ensuring the sensitivity, specificity, and reliability required in a POCT. POCT technologies offer advantages including ease of use, high diagnostic accuracy, rapid clinical assessment, and cost-effectiveness in manufacturing and consumables. Additionally, the Review highlights current challenges and future perspectives for delivering personalized healthcare through portable and wearable POCT systems that operate on a sample-in, result-out basis.

A Colorimetric Biosensor Integrating Rotifer-Mimicking Magnetic Separation with RAA/CRISPR-Cas12a for Rapid and Sensitive Detection of Salmonella

Efficient detection of foodborne bacteria is crucial for ensuring food safety, yet current methods often fall short in balancing speed, accuracy, sensitivity, and cost. This study presents an integrated biosensing platform for the rapid and sensitive detection of Salmonella in large-volume food samples. The platform incorporates a Rotifer-Mimicking Magnetic Separator (RMMS) that enhances the sample pretreatment by effectively mixing and isolating the bacteria from the sample. Coupled with this, the colorimetric biosensor utilizes a streamlined one-pot system that combines Recombinase Aided Amplification (RAA), betaine, and CRISPR-Cas12a to enable efficient pathogen detection. Initially, phenylboronic acid-modified magnetic beads (PBA-MBs) capture Salmonella, forming bacteria-PBA-MB complexes, which are then isolated using the RMMS. Target DNA amplicons activate ribonucleoprotein complexes, and Au@PtNPs-MBs with linker single DNAs are cleaved to release Au@PtNPs. The Au@PtNPs catalyze the H2O2-3,3',5,5'-tetramethylbenzidine, producing a visible blue color that indicates Salmonella concentration. This biosensor successfully detects Salmonella in 40 mL spiked milk samples within 75 min, achieving a detection limit of 89 CFU/mL. This work offers a simple, sensitive, low-cost detection method with potential applications in on-site testing, significantly enhancing food safety monitoring.

GARNN: An interpretable graph attentive recurrent neural network for predicting blood glucose levels via multivariate time series

Accurate prediction of future blood glucose (BG) levels can effectively improve BG management for people living with type 1 or 2 diabetes, thereby reducing complications and improving quality of life. The state of the art of BG prediction has been achieved by leveraging advanced deep learning methods to model multimodal data, i.e., sensor data and self-reported event data, organized as multi-variate time series (MTS). However, these methods are mostly regarded as "black boxes" and not entirely trusted by clinicians and patients. In this paper, we propose interpretable graph attentive recurrent neural networks (GARNNs) to model MTS, explaining variable contributions via summarizing variable importance and generating feature maps by graph attention mechanisms instead of post-hoc analysis. We evaluate GARNNs on four datasets, representing diverse clinical scenarios. Upon comparison with fifteen well-established baseline methods, GARNNs not only achieve the best prediction accuracy but also provide high-quality temporal interpretability, in particular for postprandial glucose levels as a result of corresponding meal intake and insulin injection. These findings underline the potential of GARNN as a robust tool for improving diabetes care, bridging the gap between deep learning technology and real-world healthcare solutions.

Deep-Learning-Assisted Microfluidic Immunoassay via Smartphone-Based Imaging Transcoding System for On-Site and Multiplexed Biosensing

Point-of-care testing (POCT) with multiplexed capability, ultrahigh sensitivity, affordable smart devices, and user-friendly operation is critically needed for clinical diagnostics and food safety. This study presents a deep-learning-assisted microfluidic immunoassay platform that uses a smartphone-based imaging transcoding system, polystyrene microsphere-based encoding, and artificial-intelligence-assisted decoding. Microspheres of varying sizes act as multiprobes, with their quantities correlating to target concentrations after an immunoreaction and separation-filtration within the microfluidic chip. A smartphone with intelligent decoding software captures images of multiprobes from the chip and performs classification, counting, and concentration calculations. The "encoding-decoding" strategy and integrated microfluidic chip design allow these processes to be completed in simple steps, eliminating the need for additional immunomagnetic separation. As a proof of concept, this platform successfully detected multiple respiratory viruses and antibiotics in various real samples with high sensitivity within 30 min, demonstrating great potential as a smart, universal toolkit for next-generation POCT applications.

Machine learning in point-of-care testing: innovations, challenges, and opportunities

The landscape of diagnostic testing is undergoing a significant transformation, driven by the integration of artificial intelligence (AI) and machine learning (ML) into decentralized, rapid, and accessible sensor platforms for point-of-care testing (POCT). The COVID-19 pandemic has accelerated the shift from centralized laboratory testing but also catalyzed the development of next-generation POCT platforms that leverage ML to enhance the accuracy, sensitivity, and overall efficiency of point-of-care sensors. This Perspective explores how ML is being embedded into various POCT modalities, including lateral flow assays, vertical flow assays, nucleic acid amplification tests, and imaging-based sensors, illustrating their impact through different applications. We also discuss several challenges, such as regulatory hurdles, reliability, and privacy concerns, that must be overcome for the widespread adoption of ML-enhanced POCT in clinical settings and provide a comprehensive overview of the current state of ML-driven POCT technologies, highlighting their potential impact in the future of healthcare.

Prediction and detection of terminal diseases using Internet of Medical Things: A review

The integration of Artificial Intelligence (AI) with the Internet of Medical Things (IoMT) has revolutionized disease prediction and detection, but challenges such as data heterogeneity, privacy concerns, and model generalizability hinder its full potential in healthcare. This review examines these challenges and evaluates the effectiveness of AI-IoMT techniques in predicting chronic and terminal diseases, including cardiovascular conditions, Alzheimer's disease, and cancers. We analyze a range of Machine Learning (ML) and Deep Learning (DL) approaches (e.g., XGBoost, Random Forest, CNN, LSTM), alongside advanced strategies like federated learning, transfer learning, and blockchain, to improve model robustness, data security, and interoperability. Findings highlight that transfer learning and ensemble methods enhance model adaptability across clinical settings, while blockchain and federated learning effectively address privacy and data standardization. Ultimately, the review emphasizes the importance of data harmonization, secure frameworks, and multi-disease models as critical research directions for scalable, comprehensive AI-IoMT solutions in healthcare.

Antimicrobial Resistance Patterns and Biofilm Analysis via Sonication in Intensive Care Unit Patients at a County Emergency Hospital in Romania

Background/Objectives: Ventilator-associated pneumonia (VAP) remains a critical challenge in ICU settings, often driven by the biofilm-mediated bacterial colonization of endotracheal tubes (ETTs). This study investigates antimicrobial resistance patterns and biofilm dynamics in ICU patients, focusing on microbial colonization and resistance trends in tracheal aspirates and endotracheal tube biofilms at a county emergency hospital in Romania. Methods: We conducted a longitudinal analysis of ICU patients requiring mechanical ventilation for more than 48 h. Tracheal aspirates and ETT biofilms were collected at three key time points: T1 (baseline), T2 (48 h post-intubation with ETT replacement), and T3 (92-100 h post-T2); these were analyzed using sonication and microbiological techniques to assess microbial colonization and antimicrobial resistance patterns. Results: In a total of 30 patients, bacteria from the ESKAPEE group (e.g., Klebsiella pneumoniae, Acinetobacter baumannii, Staphylococcus aureus) dominated the microbiota, increasing their prevalence over time. Resistance to carbapenems, colistin, and vancomycin was notably observed, particularly among K. pneumoniae and A. baumannii. Biofilm analysis revealed high persistence rates and the emergence of multidrug-resistant strains, underscoring the role of ETTs as reservoirs for resistant pathogens. The replacement of ETTs at T2 correlated with a shift in microbial composition and reduced biofilm-associated contamination. Conclusions: This study highlights the temporal evolution of antimicrobial resistance and biofilm-associated colonization in a limited number of ICU patients (30 patients). The findings support implementing routine ETT management strategies, including scheduled replacements and advanced biofilm-disruption techniques, to mitigate VAP risk and improve patient outcomes.

Rapid and comprehensive detection of viral antibodies and nucleic acids via an acoustofluidic integrated molecular diagnostics chip: AIMDx

Precise and rapid disease detection is critical for controlling infectious diseases like COVID-19. Current technologies struggle to simultaneously identify viral RNAs and host immune antibodies due to limited integration of sample preparation and detection. Here, we present acoustofluidic integrated molecular diagnostics (AIMDx) on a chip, a platform enabling high-speed, sensitive detection of viral immunoglobulins [immunoglobulin A (IgA), IgG, and IgM] and nucleic acids. AIMDx uses acoustic vortexes and Gor'kov potential wells at a 1/10,000 subwavelength scale for concurrent isolation of viruses and antibodies while excluding cells, bacteria, and large (>200 nanometers) vesicles from saliva samples. The chip facilitates on-chip viral RNA enrichment, lysis in 2 minutes, and detection via transcription loop-mediated isothermal amplification, alongside electrochemical sensing of antibodies, including mucin-masked IgA. AIMDx achieved nearly 100% recovery of viruses and antibodies, a 32-fold RNA detection improvement, and an immunity marker sensitivity of 15.6 picograms per milliliter. This breakthrough provides a transformative tool for multiplex diagnostics, enhancing early infectious disease detection.

Full Spectral Overlap to Enhanced Fluorescence Quenching Ability by Using Covalent Organic Frameworks as a Springboard of Quencher for the Turn-on Fluorescence Immunoassay

According to the fluorescence internal filtering effect (IFE), the more the absorption spectrum of the quencher overlaps with the excitation and emission spectra of the fluorescent substance, the better the quenching effect and, correspondingly, the more significant and sensitive the contrast becomes when the fluorescence is turned on. Thus, in the competitive fluorescence-quenching lateral flow immunoassays (FQ-LFIAs), the fluorescence quencher with an outstanding optical property is of great importance. Herein, gold nanoparticles (AuNPs) and polydopamine (PDA) coengineered covalent organic frameworks (COF/Au@PDA) were synthesized as a fluorescence quencher to increase spectral overlap. Thanks to the excellent visible light absorption of COF with donor-acceptor (D-A) structure, the localized surface plasmon resonance (LSPR) capability of AuNPs, and the broad light absorption of the PDA layer, the COF/Au@PDA exhibits intense absorption and a full spectral overlap toward aggregation-induced emission luminous (AIE) dots. Thereafter, COF/Au@PDA, with its immense potential to completely quench the fluorescence of AIE dots through primary IFE and secondary IFE, was applied to a bimodal LFIA platform for verification with a nitrofurazone metabolite as a model analyte. As expected, the detection sensitivity of the COF/Au@PDA-based FQ-LFIA (turn-on) is improved by 6-fold compared with that of the colorimetric (CM)-LFIA (turn-off). Further, ChatGpt was used to improve the assay accuracy and sensitivity, utilizing its high sensitivity to subtle changes in LFIA signals, especially for weak signals that are indeterminate with the naked eye. This work offers a potential approach for building a high-performance fluorescence quencher in the FQ-LFIA and indicates the potential for the application of artificial intelligence in highly sensitive LFIAs.

Machine Learning-Based Quantification of Lateral Flow Assay Using Smartphone-Captured Images

Lateral flow assay has been extensively used for at-home testing and point-of-care diagnostics in rural areas. Despite its advantages as convenient and low-cost testing, it suffers from poor quantification capacity where only yes/no or positive/negative diagnostics are achieved. In this study, machine learning and deep learning models were developed to quantify the analyte load from smartphone-captured images of the lateral flow assay test. The comparative analysis identified that random forest and convolutional neural network (CNN) models performed well in classifying the lateral flow assay results compared to other well-established machine learning models. When trained on small-size images, random forest models excelled CNN models in image classification. Contrarily, CNN models outperformed random forest models in classifying noisy images.

Unsupervised Learning-Assisted Acoustic-Driven Nano-Lens Holography for the Ultrasensitive and Amplification-Free Detection of Viable Bacteria

Bacterial infection is a crucial factor resulting in public health issues worldwide, often triggering epidemics and even fatalities. The accurate, rapid, and convenient detection of viable bacteria is an effective method for reducing infections and illness outbreaks. Here, an unsupervised learning-assisted and surface acoustic wave-interdigital transducer-driven nano-lens holography biosensing platform is developed for the ultrasensitive and amplification-free detection of viable bacteria. The monitoring device integrated with the nano-lens effect can achieve the holographic imaging of polystyrene microsphere probes in an ultra-wide field of view (∽28.28 mm2), with a sensitivity limit of as low as 99 nm. A lightweight unsupervised learning hologram processing algorithm considerably reduces training time and computing hardware requirements, without requiring datasets with manual labels. By combining phage-mediated viable bacterial DNA extraction and enhanced CRISPR-Cas12a systems, this strategy successfully achieves the ultrasensitive detection of viable Salmonella in various real samples, demonstrating enhanced accuracy validated with the qPCR benchmark method. This approach has a low cost (∽$500) and is rapid (∽1 h) and highly sensitive (∽38 CFU mL-1), allowing for the amplification-free detection of viable bacteria and emerging as a powerful tool for food safety inspection and clinical diagnosis.

Evaluation of oxidative stress biomarkers for differentiating bacterial and viral infections: a comparative study of glutathione disulfide (GSSG) and reduced glutathione (GSH)

Background and aims

This study evaluates the potential of oxidative stress biomarkers, specifically glutathione disulfide (GSSG) and reduced glutathione (GSH), for differentiating bacterial and viral infections. Oxidative stress plays a crucial role in the immune response, and glutathione is a key regulator of cellular redox balance. The aim was to assess whether differences in GSH and GSSG levels could be used as diagnostic markers for infection type.

Methods

A chemiluminescence-based method evaluated GSH and GSSG as potential biomarkers for distinguishing between bacterial and viral infections. The GSH and GSSG concentrations were analyzed across bacterial, viral, and control groups.

Results

Our data revealed significant differences in the GSH/GSSG ratio between the analyzed groups, with bacterial infections showing higher oxidative stress markers compared to viral infections. A combined analysis of GSH and GSSG concentrations, visualized through heatmaps and ROC curves, improved diagnostic accuracy, with clustering patterns distinguishing infection types.

Conclusions

These findings suggest that the GSH/GSSG ratio could be used as a biomarker in distinguishing between bacterial and viral infections, offering potential clinical applications for more accurate diagnosis. Further research is required to validate these results in larger cohorts and to explore the underlying mechanisms of oxidative stress in pathogen-specific immune responses.

Rapid and automated interpretation of CRISPR-Cas13-based lateral flow assay test results using machine learning

CRISPR-Cas-based lateral flow assays (LFAs) have emerged as a promising diagnostic tool for ultrasensitive detection of nucleic acids, offering improved speed, simplicity and cost-effectiveness compared to polymerase chain reaction (PCR)-based assays. However, visual interpretation of CRISPR-Cas-based LFA test results is prone to human error, potentially leading to false-positive or false-negative outcomes when analyzing test/control lines. To address this limitation, we have developed two neural network models: one based on a fully convolutional neural network and the other on a lightweight mobile-optimized neural network for automated interpretation of CRISPR-Cas-based LFA test results. To demonstrate proof of concept, these models were applied to interpret results from a CRISPR-Cas13-based LFA for the detection of the SARS-CoV-2 N gene, a key marker for COVID-19 infection. The models were trained, evaluated, and validated using smartphone-captured images of LFA devices in various orientations with different backgrounds, lighting conditions, and image qualities. A total of 3146 images (1569 negative, 1577 positive) captured using an iPhone 13 or Samsung Galaxy A52 Android smartphone were analyzed using the trained models, which classified the LFA results within 0.2 s with 96.5% accuracy compared to the ground truth. These results demonstrate the potential of machine learning to accurately interpret test results of CRISPR-Cas-based LFAs using smartphone-captured images in real-world settings, enabling the practical use of CRISPR-Cas-based diagnostic tools for self- and at-home testing.

Analyte-induced hindrance in the RCA-assisted CRISPR/Cas12a system for homogeneous protein assays

Heterogeneous assays, such as enzyme-linked immunosorbent assays, have become indispensable for in vitro diagnostics. However, the simple, sensitive, and accurate detection is limited by their multiple washing and incubation steps, and limited amplification methods. In this study, we design a novel approach utilizing analyte-induced hindrance within the rolling circle amplification (RCA)-assisted CRISPR/Cas12a system for simple and highly sensitive homogenous protein detection. Streptavidin (SA) and digoxin antibody (anti-Dig) are employed as representative detection models. The specific recognition of target proteins using primers modified with small molecules hinders the RCA process, preventing the activation of Cas12a's trans-cleavage activity, thereby leading to a reduction in fluorescence intensity. Our developed platform exhibites exceptional detection performance characterized by high sensitivity, robust specificity, and significant potential for application in complex samples. By expanding the recognition elements, this platform can evolve into a versatile clinical diagnostic tool with universal applicability. In addition, this platform provides a novel direction for quantifying ultralow-concentration disease biomarkers in clinical practice.

Universal All-In-One Lateral Flow Immunoassay with Triple Signal Amplification for Ultrasensitive and Simple Self-Testing of Treponema pallidum Antibodies

Lateral flow immunoassay (LFIA) is valued for its simplicity and rapidity for on-site screening, however, it experienced false negatives in real sample analysis due to low sensitivity. Although many signal amplification techniques can improve the sensitivity, they usually require additional complicated steps. To address these issues, taking Treponema pallidum (T. pallidum) antibodies as a model detecting target, herein, we report an all-in-one LFIA (AIO-LFIA) with triple-step signal amplification to significantly improve sensitivity while maintaining simplicity. This LFIA utilizes a biotin-streptavidin system for initial signal amplification, followed by introducing a release controller with a specific imprinted structure for timed multicomponent release, which avoids the extra steps when adding components in traditional LFIA. Particularly, a 3D-printed programmed metal in situ growth (MISG) device is integrated to localize signal enhancement at specific sites, overcoming limitations of traditional MISG and substantially reducing reagent usage and assay time, and the nitrocellulose membrane surface was much cleaner than the conventional approach, which facilitates signal readout. After optimization, the proposed AIO-LFIA is capable of visual detection down to 1 pg/mLT. pallidum antibodies in 15 min, 1000-fold lower than the gold nanoparticle-based LFIA. In clinical testing of 152 samples, the AIO-LFIA can distinguish all positive samples, outperforming commercial LFIA which missed those positive samples with relatively low antibody levels. Thus, this study presents a universal ultrasensitive and reliable AIO-LFIA strategy for infectious diseases self-testing, providing an effective promising prospect to address the challenge over emerging infectious diseases in the future.

AI-Reinforced Wearable Sensors and Intelligent Point-of-Care Tests

Artificial intelligence (AI) techniques offer great potential to advance point-of-care testing (POCT) and wearable sensors for personalized medicine applications. This review explores the recent advances and the transformative potential of the use of AI in improving wearables and POCT. The integration of AI significantly contributes to empowering these tools and enables continuous monitoring, real-time analysis, and rapid diagnostics, thus enhancing patient outcomes and healthcare efficiency. Wearable sensors powered by AI models offer tremendous opportunities for precise and non-invasive tracking of physiological conditions that are essential for early disease detection and personalized treatments. AI-empowered POCT facilitates rapid, accurate diagnostics, making these medical testing kits accessible and available even in resource-limited settings. This review discusses the key advances in AI applications for data processing, sensor fusion, and multivariate analytics, highlighting case examples that exhibit their impact in different medical scenarios. In addition, the challenges associated with data privacy, regulatory approvals, and technology integrations into the existing healthcare system have been overviewed. The outlook emphasizes the urgent need for continued innovation in AI-driven health technologies to overcome these challenges and to fully achieve the potential of these techniques to revolutionize personalized medicine.

From Lab to Home: Ultrasensitive Rapid Detection of SARS-CoV-2 with a Cascade CRISPR/Cas13a-Cas12a System Based Lateral Flow Assay

Currently, CRISPR/Cas-based molecular diagnostic techniques usually rely on the introduction of nucleic acid amplification to improve their sensitivity, which is usually more time-consuming, susceptible to aerosol contamination, and therefore not suitable for at-home molecular testing. In this research, we developed an advanced CRISPR/Cas13a-Cas12a-based lateral flow assay that facilitated the ultrasensitive and rapid detection of SARS-CoV-2 RNA directly from samples, without the need for nucleic acid amplification. This method was called CRISPR LFA enabling at-home RNA testing (CLEAR). CLEAR used a novel cascade mechanism with specially designed probes that fold into hairpin structures, enabling visual detection of SARS-CoV-2 sequences down to 1 aM sensitivity levels. More importantly, CLEAR had a positive coincidence rate of 100% and a negative coincidence rate of 100% for clinical nasopharyngeal swabs from 16 patients. CLEAR was particularly suitable for at-home molecular testing, providing a low-cost, user-friendly solution that can efficiently distinguish between different SARS-CoV-2 variants. CLEAR overcame the common limitations of high sensitivity and potential contamination associated with traditional PCR-based systems, making it a promising tool for widespread public health application, especially in environments with limited access to laboratory resources.

Aspects of 6th generation sensing technology: from sensing to sense

The 6th-generation (6G) sensing technology is transforming the ways we perceive and interact with the world in real scenarios. It combines advanced materials, sophisticated algorithms, and connectivity to create intelligent, context-aware systems that can interpret and respond to environmental stimuli with unprecedented accuracy and speed. The key advancements include 1) ultra-sensitive sensors capable of detecting physical, chemical, and biological changes at low concentrations, 2) the integration of artificial intelligence (AI) and machine learning (ML) for enhanced data processing, and 3) the deployment of IoT networks with 5th-generation (5G) for seamless data transmission and real-time analysis. These cutting-edge technologies create immersive environments where devices capture data and anticipate user needs and environmental conditions. The 6G sensing technology has potential applications across sectors like point-of-care (PoC), healthcare, urban planning, and environmental monitoring. The transition from sensing to sense-making represents a paradigm shift, fostering a more intuitive, responsive, and interconnected world. The article provides a comprehensive overview of the current state and prospects of 6G sensing technology, highlighting its transformative potential and the challenges in realizing its full capabilities.