The management of airway disease based on phenotypic characteristics is limited by the absence of diagnostic tools capable of concurrently addressing multiple inflammatory pathways. Traditional assessments provide isolated measurements, which fail to capture intricate interactions among early immune activation, systemic dissemination, neutrophil recruitment, and viral immune response. These are four critical mechanisms that influence disease progression and therapeutic outcomes. In this study, we introduce a non-faradaic electrochemical impedance spectroscopy biosensor designed for a label-free parallel measurement of C-reactive protein (CRP), IP-10 (CXCL10), interleukin-6 (IL-6), and interleukin-8 (IL-8). The platform uses frequency-specific impedance analysis to detect binding events in capacitance at the electrode-electrolyte interface, allowing for direct concentration-dependent detection. It offers wide dynamic ranges covering both physiological and pathological concentrations for all four analytes. Analytical validation showed a strong correlation between spiked and recovered concentrations (R2 > 0.95), with percent recovery within the 80-120% accuracy range and cross-reactivity below 20% under multiplexed conditions, confirming the platform's selectivity and quantitative reliability. Both intra- and inter-assay reproducibility met analytical standards. By incorporating four distinct inflammatory biomarkers into a single impedance framework, this platform moves airway diagnostics beyond isolated measurements to a multi-dimensional phenotype resolution, laying the groundwork for decentralized risk assessment and therapy-guided inflammatory monitoring.
Extracellular vesicles (EVs) are membrane-bound nanoparticles that serve as minimally invasive biomarkers, yet their clinical translation is hampered by limitations in existing isolation and detection strategies. Here, we systematically benchmarked EVs secreted from HeLa cell cultures using ultracentrifugation, microfluidic ultrafiltration, and magnetic bead–based capture, the latter implemented both conventionally and on OpenDrop, an off-the-shelf digital microfluidic (DMF) platform. This comparative evaluation encompassed key performance metrics, including yield, purity, workflow complexity, and operational efficiency. Nanoparticle tracking analysis and scanning electron microscopy revealed that EVs isolated from DMF exhibited recovery rates and vesicle morphology on par with those obtained through conventional bead-based workflows, while requiring markedly reduced reagent volumes. Western blotting and flow cytometry confirmed robust expression of tetraspanin EV markers (CD63 and CD81), across a dynamic concentration range. On-chip ELISA performed directly on the DMF device demonstrated strong linearity across EV concentrations ranging from 5 × 106 to 1 × 109 particles/mL (R2 = 0.8935), in close agreement with conventional benchtop measurements (R2 = 0.8388). Together, these results establish the DMF platform used in this study as a versatile, and cost-efficient solution capable of integrating EV isolation and detection within a single, unmodified device. By uniting conventional laboratory workflows with miniaturized, digitally controlled operations, this work underscores the potential of accessible DMF technologies to advance portable point-of-care EV diagnostics in the future.
Organ-on-a-chip (OoC) platforms are a key subclass of microphysiological systems (MPS), designed to mimic the dynamic and physiological properties of native human tissue and organs. These systems have attracted significant attention, particularly following the Food and Drug Administration (FDA) approval of their use in the pharmaceutical and biomedical industries as alternatives to traditional models. However, their physiological relevance is often limited due to the lack of functional immune compartments, which are responsible for maintaining homeostasis and to modulate disease progression in the human body. In this review, we present a critical overview of immune-integrated OoC platforms, highlighting their capacity to emulate complex immunological processes such as immune cell trafficking, extravasation, and cytokine-mediated inflammation. We further explore recent advances in microfluidic design and on-chip biosensor integrations to perform real-time monitoring of immune dynamics. Despite their immense potential for drug development and personalized medicine, we highlight that their clinical translation remains challenged by biological and microarchitectural limitations, the lack of systemic multi-organ networks, hurdles in on-chip biosensor integration, and operational drawbacks. Overall, we outline a strategic roadmap for the standardization of immunocompetent OoC platforms to generate the robust and reproducible datasets required for artificial intelligence (AI) models and subsequent FDA qualification.
Sensitive detection of biomolecules in complex, non-transparent biological media is highly desirable for diagnostics and food safety. Semiconductor nanowires (NW) strongly amplify fluorescence signals, however their application has been largely limited to transparent media due to the scattering and absorption arising in turbid sample matrices. We present a detection platform based on polymer-supported semiconductor NWs that enables highly sensitive fluorescent detection in turbid media using a model biotin–streptavidin system. By anchoring the NWs vertically in an optically transparent polymer, both excitation and emission are directed through the NWs rather than through the sample, thereby decoupling the optical path from the turbid medium and reducing scattering- and absorption-related light attenuation. This enables sensitive detection of fluorescent signals from nanowire-bound molecules in highly scattering fluids using standard epi-fluorescence microscopy. We demonstrate detection of fluorescently labeled streptavidin at 1 nM in reconstituted powdered milk and 100 pM in whole human blood and lipid emulsion. This method enables spatially resolved, highly sensitive measurements with minimal sample preprocessing and without relying on specialized techniques, such as TIRF.
Integration is one of the most challenging issues in Lab-on-a-Chip (LOC) devices and a major hurdle for transforming LOCs into practical applications. To date, cumbersome solutions have been proposed involving either heterogeneously integrated cartridges or various microfluidic components interconnected via various types of tubing and microfluidic fittings. In this work, we present a fully integrated LOC in which key components required to perform a complete analysis of biological samples such as heating elements, microfluidic channel networks and biosensors are seamlessly integrated on a single printed circuit board (PCB). This approach allows for the implementation of complicated functions such as nucleic acid amplification and detection on the same low cost, mass producible chip. In our realization, graphene oxide (GO) is deposited as a multilayer stack and then reduced via a mild thermal treatment on predefined Au plated electrodes as a means to straightforward implement highly sensitive biosensors on the chip. As a proof-of-concept application, a specific DNA region of BRCA1 gene is successfully amplified on-chip using recombinase polymerase amplification (RPA) and subsequently detected by the on-chip GO biosensors. The success of the DNA amplification is verified off-chip via gel electrophoresis, whereas the entire amplification and detection procedure is completed within 40 min.
Electrolyte-gated organic transistors (EGOTs) are emerging as promising devices for biosensing applications due to their high transconductance, ability to operate in liquid environments at low voltages and label-free detection capabilities. In this study, we investigate the performance of EGOTs based on the donor-acceptor polymer DPP-DTT, a material known for its high charge mobility and aqueous stability. The device was tested as a biosensor for the detection of mitochondrial DNA (mtDNA), a biomarker associated with multiple sclerosis, as a model analyte. The gate electrode was functionalized with a complementary DNA probe via surface chemistry, with functionalization efficiency validated by surface plasmon resonance and plasmon-enhanced fluorescence spectroscopy. The resulting DPP-DTT-based EGOT sensor exhibited specific detection of mtDNA in solution, with a detection range from 70 pM to 5 nM. These results highlight the potential of DPP-DTT based EGOTs as stable, sensitive platforms for biosensing applications.
Proteases are emerging biomarkers for diseases including cancer, cardiovascular disorders, and inflammation, making their detection valuable for early diagnosis and therapeutic monitoring. Translating biomarker monitoring from laboratories to personalized healthcare requires biosensors that are simple, selective, and compatible with point-of-care formats. To the best of our knowledge, we report the first peptide-based chronopotentiometric biosensor for protease detection. The sensor is fabricated on cost-effective disposable electrodes, has a straightforward design, and operates at ultra-low intrinsic sensing power (<1 nW), supporting future integration into compact sensing devices.The sensor enables direct, label-free detection of matrix metalloproteinase-9 (MMP-9) with picomolar detection limit through potential shifts induced by peptide hydrolysis. Careful peptide design provides high selectivity against other MMP family members, and the device successfully detects proteases released from cells in moderately complex media. Inhibition studies confirm that the signal originates from proteolytic activity.In parallel, the biosensing platform is validated using multi-parametric surface plasmon resonance (MP-SPR), providing an independent and robust optical method to confirm peptide–MMP-9 interactions. This work demonstrates peptide-based dual-platform sensing for highly sensitive, real-time protease detection and establishes a versatile foundation for future point-of-care and wearable biosensing applications.
Monitoring pathogenic bacteria in anthropogenic and natural water reservoirs remains a major challenge for health authorities, requiring the development of rapid, portable, low-labor, and cost-effective detection methods. Digital photocorrosion (DIP) using GaAs/AlGaAs biosensors has recently demonstrated promising results for detecting Legionella pneumophila and Escherichia coli, with increasing bacterial concentrations resulting in delayed biochip DIP rates. However, bacterial spores, such as those of the Bacillus cereus group, induced accelerated DIP rates, creating a significant challenge for the interpretation of spore biosensing responses. In this study, we investigated the mechanisms underlying the accelerated DIP rates induced by Bacillus thuringiensis subsp. kurstaki spores immobilized on biofunctionalized surfaces of GaAs-AlGaAs biochips. Fourier-transform infrared (FTIR) spectroscopy and conductivity measurements suggested that this distinctive behavior is associated with activation-related modifications of the local ionic environment surrounding viable spores. Meanwhile, heat-killed spores and vegetative cells showed delayed DIP rates, indicating that the biosensor can distinguish viable spores from inactivated spores and vegetative cells. These findings provide new mechanistic insight into biological processes influencing DIP measurements and contribute to a more reliable interpretation of bacterial spore biosensing. The results are expected to support further optimization and validation of robust, field-deployable DIP biosensors for detecting bacterial spores in environmental samples.
Cervical cancer ranks as the fourth most common cancer and is the third leading cause of cancer-related deaths among women worldwide. Globally, an estimated of 66,0000 women were diagnosed with cervical cancer, with 35,0000 deaths in 2022. Traditional cervical cancer screening techniques, including pap smear, visual inspection with acetic acid (VIA), and PCR-based HPV DNA assays, require advanced centralised laboratories, complex sample preprocessing, skilled personnel, high operational costs, and long turnaround time, thereby restricting their availability in point-of-care, limited-resource settings where extensive screening is essential. The reported electrochemical sensor overcomes the limitations of traditional methods by developing an amplification-free, sample pre-processing-free, nanoparticle-free, smartphone-integrated electrochemical sensing platform, which can detect HPV L1 protein from unprocessed cervical swab samples. The sensor exhibits high sensitivity with a detection limit of 0.29 pM in a wide detection range of 1 pM - 6 nM. Further, the aptasensor shows high reproducibility of RSD 3.85 %, selectivity, and stability up to 30 days at room temperature. All the electrochemical signal responses in the presence of cervical swab samples exhibit high correlation with conventional COBAS testing. Commercially, the fabrication cost for the disposable sensor chip is very low (0.07 USD), which is very important for population-based screening and point-of-care testing. The compatibility for large-scale fabrication and integration with portable readout systems like smartphones also broadens its market appeal in point-of-care diagnostic sectors.
Blood group determination is essential for transfusion safety and surgical planning; however, conventional serological methods are invasive and resource-dependent. This study presents an integrated bioengineering framework for non-invasive ABO/Rh phenotype prediction using dermatoglyphic fingerprint analysis and hybrid deep learning. The proposed system combines standardized optical fingerprint acquisition with computational feature extraction and classification. A dataset of 6000 publicly available fingerprint images was used for model development (85/15 training/testing split), followed by independent external validation on 300 real-world samples. Four network topologies were evaluated: ResNet-50, ResNet-101, an unweighted hybrid model, and a class-weighted hybrid model. The weighted hybrid configuration achieved the highest training accuracy (96.65%). During external validation, it demonstrated superior generalization performance with 96.73% accuracy and 10 misclassifications, whereas the unweighted hybrid achieved 94.33% accuracy. Statistical assessment using 95% Wilson confidence intervals and two-proportion z-tests confirmed significant improvements over single-backbone architectures (p < 0.001). These findings support the feasibility of a fingerprint-based predictive framework as a rapid, non-invasive screening and decision-support tool for preliminary assessment in emergency and resource-limited healthcare settings.
We present a method for enhancing fluorescence signals in centrifugal microfluidic systems by integrating a plano-convex lens into the optical path of a fluorescence detector, effectively focusing emitted light onto a single point. Two-dimensional computational simulations were first conducted to optimize the lens design and positioning, resulting in a 1.7-fold increase in ray collection efficiency compared to configurations without the lens. When the optimally designed plano-convex lens was incorporated into the three-dimensional detection chamber of an actual centrifugal microfluidic system, fluorescence intensities from fluorescein solutions increased by approximately 40–80-fold across most of the tested concentration range. Finally, we applied this strategy to enhance fluorescence signals generated by reverse transcription loop-mediated isothermal amplification (RT-LAMP) assays targeting Influenza B virus, where the lens-integrated disc achieved approximately 60-fold higher fluorescence signals than the lens-free disc. These findings indicate that this strategy can substantially improve fluorescence detection sensitivity in assays employing similar optical configurations and detection principles.
Polyaniline (PANI)-based materials continue to bridge materials science across agricultural electronics due to their inherent redox behavior, structural tunability, and ease of chemical synthesis. However, translating lab-scale polymer performance into resilient, real-world field diagnostics requires overcoming multi-interfacial degradation, thermal and humidity cross-sensitivity, and unoptimized material combinatorics. Unlike prior reviews focusing on narrow aspects of conducting polymers, this cross-cutting work synthesizes chemical oxidative, electrochemical, and green synthesis routes for PANI, critically evaluating how acid dopants and nanomaterial fillers, like MXenes, graphene and metal oxides, govern electrical and mechanical transport properties. Emerging fabrication strategies like inkjet printing, dip-coating, and laser-induced graphene are highlighted for scalable deployment. Moving beyond isolated applications, this review introduces a novel material-to-algorithm decision superstructure designed to match specific agronomic target analytes with optimal multi-layer sensing, substrate, and encapsulation configurations. Crucially, we elucidate the integration of physics-informed artificial intelligence pathways into the sensor lifecycle by separating operations into AI for signal processing and AI for material design optimization. The framework utilizes Particle Swarm Optimization and Non-dominated Sorting Genetic Algorithm II to navigate non-linear compositional spaces, while implementing temporal Long Short-Term Memory networks and goal-oriented Agentic AI loops for advanced field signal processing. To ensure field reliability, post-hoc Explainable AI algorithms are introduced to mathematically audit sensor decisions against true biochemical interactions, mitigating black-box shortcuts of traditional AI. Finally, we analyze systemic edge-computing and material bottlenecks under field stresses, providing an integrated engineering roadmap to exploit ternary PANI nanocomposite synergies for autonomous, self-correcting, and scientifically validated next-generation smart agricultural sensors.
Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and the progressive version Metabolic Dysfunction-Associated Steatohepatitis (MASH) are rising rapidly, yet most are still diagnosed late, because a liver biopsy is invasive and elastography and other specialist tools are not everywhere available. This is the reason why the time is now, since low-cost point-of-care testing (PoCT) technologies, smartphone-based optics, electrochemical sensors, lab-on-a-chip technologies, spectroscopy, breath/VOC, and Artificial Intelligence & Machine Learning (AI/ML) can take the noisy, variable signals produced by these technologies and make them reliable, repeatable signals. What is new in this article is that instead of discussing PoCT technologies as such, and instead of listing biomarkers in isolation, we are going to introduce the reader to the concept of a “biology → measurables → actions” framework in the context of MASLD/MASH management and explain how AI/ML needs to work throughout the entire PoCT process, from signal cleaning and automated quality control, to calibration transfer, uncertainty-based reporting, and decision support. This signal-to-decision approach distinguishes this review from the majority of existing articles, which focus mainly on the device's hardware, a single marker, or the accuracy of the most recent models but fail to consider the constraints of deployment. We will consider the major PoCT modality families of interest for NAFLD/NASH, including optical (colorimetry, fluorescence, chemiluminescence, electrochemiluminescence), electrochemical, microfluidics-based lab-on-chip, spectroscopy (Raman/SERS, NIR), and breath sensors/VOC sensor arrays. Clinicians, primary-care systems, researchers, and device developers will benefit from this roadmap to design safe, scalable MASLD/MASH screening and monitoring tools.
Electric Cell-substrate Impedance Sensing (ECIS) is widely applied as a label-free and non-invasive technology for the in vitro monitoring of tissue dynamics through various electrical parameters. Fitting those to circuit models yields insights about cell-substrate and cell-cell interactions. This work reports progress in monitoring complex biological systems with ECIS technology, i.e. cell-invasion dynamics, and explores the possibility of quantifying the invasion progress over time.A human in vitro implantation model was utilized, which allowed to examine the interaction between embryonic trophoblast cells and the uterine endometrial epithelium. It involves attachment of the trophoblast to the endometrium followed by invasion. The process was evaluated by monitoring impedance amplitude and phase. Based on cut-off frequencies, quantification of cell resistance and capacitance was achieved by decomposing the impedance spectra to track the invasion process at various time points. To better understand trophoblast invasion dynamics, the in vitro equivalent electric circuit (EEC) model from conventional ECIS theory was extended to a three-component model considering the electrode, endometrial monolayer and trophoblast spheroid. Taken together, a multi-step invasion process is proposed: 1) preinvasion - non-contact condition of the spheroidal trophoblast with the endometrial monolayer, 2) early-stage invasion (0-6 h) - trophoblast attachment and endometrial penetration, and 3) late-stage invasion (6-48 h) - spheroid flattening and expansion within the endometrial monolayer.This work advances the possibility of cell-electrical quantification to monitor biological processes which involve cell and tissue intercalation. This is not only relevant for embryo implantation but also for developmental processes, inflammation and tumor invasion.
Pancreatic cancer is one of the most lethal malignancies, largely due to late-stage diagnosis and the absence of accessible screening tools. Conventional diagnostic approaches, including imaging and biopsy, are costly, invasive, and unsuitable for routine monitoring, while standard biochemical assays remain time-consuming and laboratory dependent. Therefore, rapid, sensitive, and non-invasive detection of pancreatic cancer biomarkers is critically needed. Here, we report a nanoengineered electrochemical aptasensor integrated with an automated urine analysis platform (“UroSmart”) for point-of-care (PoC) detection of carbohydrate antigen 19.9 (CA19.9). The UroSmart system incorporates a closed-loop, urine collection module and precisely controlled microfluidic dispensing system, facilitating seamless aseptic integration of sample collection, processing, and analysis within a single platform. The sensing interface uses an aptamer on gold nanoparticle-modified electrodes to improve target recognition and signal transduction. Upon sample introduction, CA19.9 is selectively captured at the electrode surface, leading to a measurable decrease in interfacial conductivity for quantitative electrochemical readout. Under optimised conditions, the platform exhibits high specificity, a wide dynamic range, and stable performance across varying physiological conditions. The limit of detection (LoD) was 0.47 U mL−1 in buffer and 2.02 U mL−1 in human urine, with a broad linear dynamic range spanning 0.1–1000 U mL−1.The integrated system enables a novel rapid, sample-in-result-out analysis, demonstrating strong potential as a reliable PoC tool for early detection and longitudinal monitoring of pancreatic cancer at the asymptomatic stage.
The use of whole cancer cells as a biomarker, as opposed to smaller analytes such as proteins, is essential for advancing diagnostic tools. In this work, a quasi-random interferometer-based optical fiber biosensor is proposed for the quantitative, label-free, and real-time detection of CD44-expressing cancer cells. The sensor is low-cost, easy to fabricate, and able to distinguish cancer cells from normal cells based on CD44 expression levels and local refractive index changes. A weak interferometric cavity at the fiber tip is responsible for the quasi-random reflection pattern, where refractive index changes induced by cell binding modulate the reflected signal in concentration-dependent manner. Quantitative analysis demonstrated a limit of detection of 48.8 cells/mL and a sensitivity of 1.1 dB/10x for the breast cancer HCC1806 cell line with high level of CD44 expression. The sensor successfully stratified cells from breast and colorectal cancers and demonstrated high specificity by clearly distinguishing cancerous cells from normal kidney epithelial cells. This work contributes to the development of biosensing platforms for whole-cell detection based on phenotypic markers, offering a simple and reproducible approach suitable for integration into microfluidic and future clinical diagnostic systems.
Cellophane is an attractive low-cost substrate for environmentally sustainable fluidic devices because it is biodegradable. Here, we report a laser direct-write (LDW) method, which relies on the principle of photo-polymerisation of a light-sensitive polymer for rapid fabrication of cellophane-based fluidic devices. In this method, a photo-polymer is first deposited on a cellophane substrate in a user-defined pattern and subsequently illuminated using a 405 nm laser to photo-polymerise the pre-deposited patterns and form hydrophobic polymer structures that form the walls of open wells and enclosed flow channels. By tuning the photo-polymer deposition speeds and deposition–photo-polymerisation cycles, polymerised structures with widths of 0.5–2 mm and heights of 0.1–2 mm were reproducibly fabricated on both uncoated and polymer-coated cellophane. Enclosed flow channels with coated cellophane as both layers supported pump-driven flow at 50–250 μL min−1, while hybrid channels with an uncoated bottom layer and coated top layer enabled pump-free capillary transport. Quantitative colourimetric detection of glucose (GOx/HRP–o-dianisidine assay) and nitrite (Griess reaction) was implemented within open wells. The limit-of-detection and limit-of-quantification were 7.31/23.8 μg mL−1 and 6.57/21.9 μM across the ranges of 10–100 μg mL−1 and 5–200 μM for glucose and nitrite respectively. We also report the detection of glucose and nitrite in artificial urine samples within both open well and enclosed flow channels. This LDW-based fabrication approach provides a practical and flexible route for rapid prototyping of disposable cellophane-based fluidic devices for colourimetric sensing.
Nanofibrous collectors made of polyvinyl butyral (PVB) and polycaprolactone (PCL) polymers were developed and evaluated as a novel alternative for forensic odor trace collection. Laboratory experiments revealed a significantly higher sensitivity of nanofibrous collectors to human odor imprints compared with currently used textile materials, as well as distinct adsorption behaviors of nanofibers made from different polymers toward molecules of diverse origin. In addition, nanofibrous collectors can be simultaneously utilized for both olfactory (canine-based) and instrumental (olfactronic) analyses. The application of nanofibrous collectors in combination with simultaneous EEG signal monitoring of working dogs may contribute to improving the objectivity and evidential reliability of canine olfactory identification, although further validation is required before such methods can be considered for legal use.
A novel and environmentally sustainable approach was used to fabricate carbon paste electrodes for insulin determination. In this work fully eco-friendly components for electrochemical sensors for insulin determination were used. The carbon material was derived from waste processing, while other components, including coconut oil as a binder and a pencil lead as a conductive contact for carbon paste electrode, were chosen for their eco-friendly nature. The prepared electrode was subsequently modified with copper via two simple methods, yielding two distinct electrode materials: a Cu-modified electrode (using cyclic voltammetry method) and a Cu2O-modified electrode (using pulse deposition method). Both materials were electrochemically studied and compared for insulin detection. It was found that the electrochemical performance was limited by different factors - adsorption for CV-Cu/CPE and charge transfer kinetics for 15pCu/CPE. Notably, the formation of a specific Cu(II)-insulin complex at the CV-Cu/CPE electrode surface led to superior performance. Linear range of insulin concentrations was within 5 nM to 4 μM, with the sensitivity 1.13 mA/μM and a low detection limit (LOD) of 5 nM in PBS, while a LOD of 0.7 μM was achieved in blood serum. This sensor demonstrated excellent selectivity in the presence of common electrochemically active species and achieved high reproducibility, with a relative standard deviation (RSD) below 10%. The results suggest that this sustainably fabricated CV-Cu/CPE sensor is a promising and effective tool for insulin detection also in real samples.
