Early cancer screening has become a critical global public health priority and necessitates highly sensitive, specific, and field-deployable bioassay technologies. Electrochemical biosensors (EBs) have shown considerable promise for cancer biomarker analysis due to their high sensitivity, rapid response, and portability, yet their clinical translation remains constrained by insufficient signal amplification and nonspecific interference in complex biological matrices. Nanozymes have emerged as a powerful solution to these challenges. This review provides an engineering-oriented framework that integrates nanozyme catalytic mechanisms with electrode design, signal transduction, and system-level integration. We examine representative applications across five biomarker categories (proteins, nucleic acids, metabolites, circulating tumor cells (CTCs), and exosomes), and highlight how detection strategies diverge fundamentally among affinity-based recognition, programmable nucleic acid amplification, direct catalytic sensing, and cell-capture formats. Key signal amplification paradigms are systematically analyzed, including direct substrate conversion, cascade systems combining nanozymes with natural enzymes or nucleic acid circuits, and relay amplification via redox mediators. We further identify persistent barriers to clinical adoption, such as batch-to-batch reproducibility, specificity in complex biological fluids, and inadequate clinical validation, and discuss engineering countermeasures from material functionalization and device integration perspectives. Finally, we outline future directions toward intelligent microsystems, wearable sensing platforms, and system integration. Collectively, this review offers a systematic engineering blueprint for transitioning nanozyme-enhanced EBs from laboratory research toward clinical cancer diagnostics.
Nanostructure-based modified electrochemical sensors have gained considerable attention for portable and non-enzymatic glucose detection, owing to their high sensitivity, operational simplicity, and elimination of enzyme-related instability. In this study, sodium dodecyl sulfate soft-template mediated synthesis of polyaniline (PANI) nanosheets and subsequent anchoring of copper oxide (CuO) nanoparticles were performed to obtain PANI–CuO (PCuO) nanocomposite. X-ray diffraction investigation revealed crystalline and semi-crystalline nature of CuO and PANI, respectively. Microscopic and energy-dispersive X-ray spectroscopy examinations demonstrated uniform distribution of CuO and sheet-like morphology of PCuO. PANI nanosheets and PCuO nanocomposites were employed to modify nickel foam for developing PANI/NF and PCuO/NF sensors and applied for the determination of glucose. The cyclic voltammogram (CV) of PCuO/NF electrode displayed characteristic redox characteristics of glucose in 0.1 M NaOH within the potential limit of −0.2 to +0.8 V. The amperometric I-t response of PCuO/NF electrode exhibited superior electrocatalytic response towards non-enzymatic glucose sensing by achieving an excellent sensitivity of 1118 µA mM⁻1 cm⁻2 and a better limit of detection of 27.0 µM. Additionally, the present sensor showed an appreciable selectivity among common physiological interferents and also unveiled good reproducibility, repeatability and long-term stability for glucose monitoring. These results highlight that PCuO/NF electrode can be utilized as a promising candidate towards non-enzymatic glucose biosensing applications.
Plant biosensors combining biological recognition elements with electrochemical, optical and CRISPR based transduction platforms have emerged as robust field deployable tools for the detection of stress biomarkers, pathogens and phytohormones directly within agricultural settings. Sophisticated computational methods are needed to tackle the potential of these devices which generate complex high dimensional data. By following PRISMA guidelines, this systematic review critically assesses the state of the art in machine learning algorithms like deep learning, random forest and time series models. These are used to convert raw biosensor data into actionable agronomic insights that provide early stress prediction, multi analyte classification and real time crop health assessment. The review provides a thorough survey of biosensing platforms including electrochemical, fluorescent, aptamer based, antibody based and multiplexed architectures as well as evaluates their coupling with Internet of Things (IoT) networks and artificial intelligence-based decision systems to enable precision agriculture. Some of the primary challenges are the lack of labelled field datasets, the complexity of making cross-crop models work for all crops and the lack of consistent machine learning benchmarking pipelines for biosensor data, though strategies such as transfer learning, self-supervised learning and data augmentation offer promising paths forward. The review discusses about future chances to put lightweight machine learning models directly into biosensor platforms for edge computing and wearables. This work is in line with the UN SDGs, especially SDG 2, SDG 12 and SDG 13 by integrating biosensor innovation with data-driven analytics for sustainable, climate-resilient agriculture.
Surface acoustic wave (SAW) sensors offer high-frequency piezoelectric transduction, rapid response, miniaturization, and potential passive or wireless operation, making them attractive for biomedical sensing. However, their practical performance is affected by complex biological matrices, environmental drift, cross-sensitivity, device variation, and weak or nonstationary signals. This review links the physical sensing mechanisms and intrinsic performance of SAW devices with the incremental contributions of artificial intelligence (AI). SAW performance is summarized for disease-biomarker detection, physiological monitoring, and human-motion sensing, while AI-assisted approaches are evaluated in terms of device optimization, drift and interference compensation, multidimensional feature extraction, and sensing-computing integration. AI can improve information yield and system robustness without replacing the underlying piezoelectric transduction or sensing interface. Key challenges include limited datasets, calibration transfer, model generalization, biofouling, matrix shifts, computational constraints, and insufficient independent or clinical validation. Future development will rely on physics-informed design, digital twins, robust biointerfaces, adaptive multimodal compensation, edge intelligence, and standardized validation.
Soft strain sensors offer compliant, lightweight, and low-cost solutions for wearable sensing and human–machine interfaces. However, their long-term performance is often degraded by baseline drift and nonlinear hysteresis caused by the viscoelastic behavior and aging of soft materials. Existing compensation approaches commonly rely on recurrent or adaptive learning models that require substantial computational resources and long historical data. Here, we present a lightweight framework for real-time compensation of time-varying errors in soft strain sensors. The framework combines an exponential moving average (EMA) algorithm for continuous drift tracking and baseline stabilization with a compact feedforward neural network (FNN) for residual nonlinearity and hysteresis compensation. Using a hydraulic filament sensor subjected to randomized cyclic stretching over two hours, the proposed method achieved a displacement prediction root-mean-square error (RMSE) of 1.18 mm (2.95% of the operating range) and a coefficient of determination (R²) of 0.99. The framework also reduced hysteresis errors from 5.76% to 1.84%. Compared with recurrent and attention-based neural network architectures, the proposed approach achieved superior accuracy while maintaining low computational complexity, making it suitable for real-time deployment. As a proof of concept, the framework was integrated into a wearable knee-motion monitoring system, achieving a knee-angle estimation RMSE of 3.63°. These results demonstrate an effective and scalable strategy for improving the long-term reliability of soft strain sensors.
Optical fiber biosensors have gained prominence as versatile platforms for label-free, real-time biochemical analysis. Among the various architectures, Tapered Optical Fibers (TOFs) and Tilted Fiber Bragg Gratings (TFBGs) stand out for their exceptional sensitivity and clinical adaptability. This review provides a comprehensive, comparative overview of these two approaches, examining their fundamental sensing mechanisms and recent application trends in detecting targets ranging from small protein biomarkers to massive, folded viral genomes. Particular attention is given to advances in plasmonic coupling, complex nanomaterial amplification, and microfluidic multiplexing. Furthermore, we critically analyzed the fundamental physical limitations that currently hinder broader commercial and in vivo deployment, specifically the shallow penetration depth of TFBGs, the environmental cross-sensitivity of TOFs, and the geometric reproducibility of both. Finally, we propose future perspectives in which the physical hybridization of tapered fibers with TFBGs, enabled by advanced fabrication techniques like post-taper femtosecond inscription, may address these bottlenecks. By combining such structural innovations with algorithmic data processing, hybrid architectures represent a promising direction for the next generation of robust, point-of-care biosensing technologies.
Micro- and nanoplastics (MNPs) in aqueous systems are increasingly recognized as analytically elusive targets, yet progress in their detection depends on more than instrumental sensitivity alone. While most existing literature remains heavily technique-focused, these accounts often underemphasize surface and interface phenomena as the root cause of method variability in real matrices. Addressing this core gap, this review suggests that MNP sensing is fundamentally an interfacial-chemistry problem: the performance of probes, capture materials, and transducers is dictated by how they engage with plastic surfaces and with the dynamic molecular layers that form around them in water. Hydrophobic, electrostatic, aromatic, and hydrogen-bonding interactions govern recognition events, while weathering, oxidation, biofouling, and eco-corona formation continuously reshape surface charge, polarity, accessibility, and signal response under realistic conditions. From this perspective, the challenge is not merely to detect particles, but to determine which particle features remain chemically accessible and analytically defensible as plastics evolve in aquatic environments. Building on this premise, the review develops a sensor-oriented framework that links particle generation, matrix effects, recognition chemistry, and signal stability to the analytical outputs being pursued, and examines how the field is moving beyond isolated techniques toward coupled and workflow-integrated systems that combine enrichment, fractionation, and orthogonal readouts. Attention is also given to early signs of real-world translation, including emerging start-up activity around rapid screening concepts, enabling chemistries, automation, and application-oriented platforms. Ultimately, mastering MNP interfacial chemistry will be decisive for turning promising detection concepts into sensors that remain selective, robust, and credible in real aqueous systems.
Flexible sensors have attracted increasing attention in motion analysis and rehabilitation medicine owing to their lightweight design, stretchability, conformability, mechanical compliance, and excellent adaptability to human interfaces. This review focuses on the application of flexible sensors in biomechanical monitoring. It systematically summarizes the fundamental principles, detectable signal types, material systems, and structural design strategies of piezoresistive, capacitive, piezoelectric, triboelectric, and other emerging flexible sensors. In addition, methodological procedures for biomechanical signal acquisition, calibration, preprocessing, feature extraction, and intelligent modeling are discussed. In motion analysis, flexible sensors can be integrated into insoles, joint patches, garments, and sports equipment for gait analysis, plantar pressure monitoring, joint motion recognition, and sports performance assessment, with machine learning further enabling motion classification and movement-state identification. In rehabilitation medicine, flexible sensors can be used to monitor joint activity, human-machine interaction pressure, plantar loading, surface electromyography, force myography, and muscle fatigue during neurorehabilitation and muscle rehabilitation, thereby providing objective evidence for rehabilitation assessment, training feedback, and individualized intervention. Despite substantial progress, challenges remain in sensing accuracy and stability, long-term wearing reliability, inter-individual variability, multisensor fusion, data standardization, and clinical validation. In the future, the integration of flexible sensors with artificial intelligence, digital twins, bioelectronics, and self-healing materials is expected to promote the development of intelligent, continuous, and clinically translatable systems for sports health monitoring and precision rehabilitation.
Electrochemical biosensors involving dehydrogenase enzymes have attracted significant attention for their capacity to enable sensitive, rapid, and cost-effective detection of therapeutically and environmentally important analytes. This paper provides a comprehensive analysis of the fundamental principles of electrochemical sensing, focusing on dehydrogenase-based systems, including their reaction mechanisms, cofactor dependence, and signal transduction pathways. Conventional detection technologies, such as optical, chromatographic, and mass spectrometry methods, are meticulously evaluated to highlight their limitations in terms of complexity, cost, and applicability to real samples. Recent advances in nanostructured materials, hybrid interfaces, and electrode engineering are systematically evaluated, demonstrating how material selection influences enzyme immobilisation, electron transfer, sensitivity, selectivity, and operational stability. The enzyme–material contact is emphasised as a critical factor influencing sensor functionality. The difficulties of enzyme immobilisation, stability, and interfacial electron transport are meticulously examined to clarify the gap between laboratory performance and real-world applicability. Additionally, issues of material toxicity, environmental impact, and sustainability are taken into account. Ultimately, translational issues such as reproducibility, fouling, scalability, and lack of standardisation are analysed to provide a pragmatic perspective on practical application. This analysis underscores the imperative for systematic interface engineering and standardised evaluation procedures to facilitate the development of reliable dehydrogenase-based electrochemical biosensors for practical applications.
Diabetes, cancer, and cardiovascular diseases present significant challenges to global healthcare systems. Early diagnosis and intervention are essential for their successful management and treatment. The use of conventional disposable biosensors can be problematic due to their limited ability and cost implications, thus resulting in the design of reusable biosensors, which offer increased durability, sustainability, and efficient monitoring of chronic disease management and drug therapy. Reusable biosensors are based on the detection-renewal-reuse processes in order to provide long-term utility in managing chronic diseases and monitoring drug therapies. In this review paper, technologies associated with the reuse of biosensors, including advanced materials, surface modifications of 2D materials, and nanotechnology, are discussed. Moreover, the regenerative processes used for biosensors, including chemical, electrochemical, physical, and biological approaches that enable sustained functionalities and drug delivery monitoring, are summarized. The application areas of these biosensors will include chronic disease detection with a focus on cancer, diabetes, and cardiovascular disease; the role of such devices in pharmacokinetic and pharmacodynamics studies will also be considered. The potential application of artificial intelligence in sensor optimization and multifunctional detection platforms is discussed as a promising approach to enhance reusable biosensor performance in personalized medicine and precision therapeutics.
The development of efficient red-emitting nanomaterials is critical for next-generation applications in bioimaging, drug delivery, and photonics. Calcium silicate nanophosphors doped with europium ions (Eu3+) are promising due to their biocompatibility and tunable optical properties. In this study, Eu3+-doped calcium silicate (Ca2SiO4:Eu3+) nanoparticles were synthesized using dendritic mesoporous silica nanoparticles (DMSNs) as structural templates. By modulating the pore architecture via ammonium nitrate etching, a series of nanostructures (C2S1-C2S5) with varying morphologies and porosities was obtained. Structural characterization confirmed the successful replication of the dendritic framework and uniform Eu3+ incorporation, particularly in C2S5. Photoluminescence analysis revealed intense red emission, with C2S5 exhibiting the highest quantum yield (81%) under 393 nm excitation. The enhanced luminescence was attributed to increased pore complexity and local structural asymmetry. Biodegradability studies in simulated body fluid showed controlled degradation, while in vitro assays confirmed efficient cellular uptake and high cytocompatibility (cell viability >75%). These results position C2S5 as a promising multifunctional nanoplatform with high luminescence efficiency, structural tunability, and excellent bioresponse, suitable for theranostic and optical applications.
Diabetes is a growing global health burden, with projections indicating that up to one in eight adults may be affected by 2050. Effective management increasingly relies on continuous glucose monitoring (CGM), yet most currently deployed systems still track a single analyte, providing limited insight into the complex metabolic perturbations that precede complications. Here, we review emerging multi-analyte and multi-modal biosensing platforms that extend CGM beyond glucose to include ketones, lactate, insulin, cortisol, electrolytes, and other biomarkers relevant to diabetes and its comorbidities. We summarize recent advances in plasmonic, electrochemical, and optical sensors and their implementation in non- or minimally invasive formats using skin, sweat, tears, and breath. We then discuss how artificial intelligence and machine learning are being integrated with these biosensors to enable predictive modeling, digital twins, and smart closed-loop insulin delivery. Emphasis is placed on affordability, scalable manufacturing, and interoperability with smartphones and telemedicine infrastructures, which are essential for equitable deployment in low- and middle-income settings. Rather than providing an exhaustive catalog, this review critically evaluates the strengths, limitations, and translational readiness of current technologies, and highlights design principles for next-generation intelligent, multi-analyte biosensing systems that can support proactive and personalized diabetes care.
We present a novel SERS-based biosensing methodology for selective detection of proteins that combines plasmonic activity and molecular recognition. Detection is performed using a surface-based plasmonic sensor featuring electrochemically deposited core-shell Ag-Au dendritic nanostructures that are chemically stable and amenable to biofunctionalization. Target selectivity to the sensor is imparted by means of antibodies immobilized on the surface of these SERS-active nanostructures. Moreover, SERS signal amplification is uniquely achieved by superimposing silver nanoparticles onto the captured analyte with the aid of an electric field. We find that such ”sandwich” architecture favors SERS signal enhancement from the top molecular layer. The sensor’s ability for sensitive and specific detection is assessed in physiologically relevant (HSA-spiked artificial urine) and complex biological media (calf serum) by employing green fluorescent protein (GFP) as the model analyte. Our results pave the way for the development of a new class of hybrid surface-based biosensors (SERS-based immunosensors) that permit sensitive detection of target proteins in complex biological matrices.
The development of simple sensing devices with rapid analytical readouts is important for practical metabolite analysis, yet many conventional assays rely on enzymes or separate testing formats. Herein, we developed a transparent-film dual-mode sensing platform integrating enzyme-free electrochemical lactate detection with semiquantitative colorimetric cholesterol assessment. The lactate-sensing zone employed a screen-printed graphene electrode modified with palladium nanoparticles (PdNPs) and activated by wide-potential-window cycling. The activated PdNP/graphene interface enhanced charge transfer and enabled lactate detection through a plausible lactate-dependent interfacial process potentially involving modulation of Pd0/PdHx-related behavior. A spatially separated ferric chloride/sulfuric acid-based colorimetric zone enabled cholesterol assessment within the same platform while preventing interference between the chemically incompatible assays. Under optimized conditions, the lactate sensor exhibited two linear ranges of 0.1–2 and 2–8 mM, with a detection limit of 82 µM. The cholesterol module provided a linear response from 10 to 500 µM within 1 min, with a detection limit of 10 µM. Analysis of spiked human serum samples yielded recoveries of 97.70–104.01% for lactate and 92.34–104.03% for cholesterol. The proposed platform provides a simple strategy for integrating dual-mode, enzyme-free metabolite analysis.
In the vast panorama of optical-based detection strategies, photoacoustic spectroscopy is particularly suitable for its versatility, high sensitivity, and scalability potential. In cantilever-enhanced photoacoustic sensors, the interferometric readout of the membrane oscillation guarantees exceptional signal-to-noise ratio, often at the expense of limited compactness. In this work, we propose a novel approach combining cantilever-enhanced photoacoustic trace-gas excitation with self-mixing interferometry for absorption signal readout. The sensor performance has been compared in terms of detection sensitivity and stability over time to that achieved with a more complex state-of-the-art readout system using a bulky balanced Michelson interferometer. Both sensors, operated with the same excitation laser emitting at 4.57µm and addressing the same target N2O line, demonstrated the same spectroscopic results in terms of signal-to-noise ratio of the acquired spectra and the same minimum detection limit of 90 parts-per-billion at tens of seconds integration time. The self-mixing readout benefits from a much reduced size, paving the way for future system downsizing and easier integration while maintaining high-sensitivity detection levels. Moreover, the intrinsic wavelength independence of photoacoustic spectroscopy, together with the broad-spectral adaptability of self-mixing readout, allows, in principle, a wide wavelength-tailorability of the sensor.
By enabling the detection of antibodies, antigens, and metabolites, immunosensing plays a central role in diagnosing and monitoring a wide spectrum of infectious and non-infectious diseases, including avian influenza virus (AIV), post-acute sequelae of SARS-CoV-2 (PASC), and cancer. However, due to the limited analytical performance of portable alternatives, quantitative immunosensing presently depends on costly, centralized laboratory equipment operated by specialists and incurs substantial ongoing expenses for reagents, maintenance, and specimen transport. Here, we present a portable, low-cost, and fully quantitative fluorescence-based immunosensing platform that delivers the performance of established laboratory immunoassays. It can quantify antibodies, antigens, and compatible metabolites in biofluids, while requiring minimal infrastructure, sample volume, and reagents. A coordinated set of innovations enabled these outcomes: hardware advances enabled miniaturized optics and reduced the sample and reagent volumes required per assay, while software advances boosted assay sensitivity and accuracy. We report the first 3D-printed polymer optical housing for fluorescence imaging, and are the first to apply log-logistic probabilistic models to separate fluorescence signals from noise. The device was validated by quantifying anti-AIV antibodies (a biomarker for tracing AIV outbreaks) and anti-SARS-CoV-2 antibodies (a biomarker for monitoring PASC) in avian and human serum, and it has participated in preliminary indoor field tests. Both assays showed performance comparable to or exceeding the published values for commercial assays, while costing less than $0.50 USD per test. By enabling cost-effective point-of-care quantitative immunosensing, this device can potentially monitor disease outbreaks, assess immunity, and improve prognosis and treatment across human and animal diseases.
Owing to significant interpatient differences in anticancer agent sensitivity and intratumoral heterogeneity at the single-cell level, rapid evaluation and selection of anticancer agents tailored to individual patients are desirable. In this study, we propose a two-stage microchamber array chip (TMC) for high-density single-cell trapping, clonal culture, and functional drug-response observation. The proposed geometry was designed to balance efficient single-cell occupancy, clonal expansion, and suppression of overproliferation within densely arranged microchambers. Comparative evaluation of multiple microchamber geometries demonstrated that the two-stage structure improved the generation of independent single-cell-derived colonies while maintaining practical single-cell occupancy. As a proof-of-concept application, human lung cancer PC9 cells were cultured in media containing different concentrations of gefitinib. The proportion of proliferating monoclonal colonies decreased with increasing anticancer agent concentration, demonstrating the feasibility of observing drug-response-associated clonal behavior within the proposed platform. These results establish the utility of the TMC as a single-cell culture platform and suggest its potential applicability to future single-cell drug-response studies.
Apurinic/apyrimidinic endonuclease 1 (APE1) is a key DNA repair enzyme and an important biomarker associated with genome stability and disease progression. In this work, we developed a dual-mode biosensor for sensitive and self-validated detection of APE1 by integrating an APE1-responsive cascade amplification module with a perovskite–Au/MXene (PAM)-supported electrochemiluminescence/surface-enhanced Raman scattering (ECL/SERS) transduction interface. In the upstream process, APE1 specifically cleaved the AP sites within an X-shaped DNA recognizer, generating trigger strands that initiated strand displacement amplification (SDA) and Zn²⁺-assisted DNAzyme cleavage, thereby releasing abundant enhancer strands. In the downstream process, the generated enhancer strands were captured by the PAM-modified electrode and further induced interfacial recruitment of reporter units, which simultaneously suppressed the ECL signal and enhanced the SERS response. As a result, APE1 activity was converted into an opposite-response dual-mode output with improved analytical reliability. Under the optimized conditions, the proposed biosensor exhibited a wide detection range from 0.005 to 2.0 U/mL and a low detection limit of 4.91 × 10⁻³ U/mL. In addition, the sensing platform showed excellent selectivity toward APE1 over various interfering nucleases and proteins, together with good repeatability, fabrication reproducibility, operational stability, storage stability, and favorable spatial uniformity of the SERS-active interface. By coupling efficient target-triggered nucleic-acid amplification with interface-confined dual-mode signal transduction, this work provides a promising strategy for sensitive and self-validated analysis of APE1 and offers a useful framework for the construction of dual-mode biosensors for other enzymatic biomarkers.
Fast and highly sensitive identification of viral antigens is essential for prompt diagnosis and efficient outbreak management. However, traditional gold nanoparticle-based immunochromatographic assays (ICAs) face limitations due to low detection sensitivity. In this study, a multifunctional molybdenum disulfide-gold nanosheet (MoS2@Au NSs) was designed as a nanoprobe and used as a signal tag for immunochromatographic test strips. This platform combines surface-enhanced Raman scattering (SERS) detection, photothermal properties, and visual colorimetric readout. The MoS2@Au NSs exhibits a remarkable photothermal conversion efficiency of 48.3%. Based on these properties, we developed a multimodal ICA (M-ICA) platform for monkeypox virus (MPXV) detection, offering a three-cooperative signal readout methods. The visual limit of detection (LOD) of the M-ICA platform for the MPXV antigen (A29L) was 0.5 ng/mL, while the SERS and photothermal modes achieved LODs of 4.2 pg/mL and 3.0 pg/mL, respectively—representing 238-fold and 333-fold improvement over conventional Au-based ICA. Furthermore, the practical utility of the platform was evaluated using inactivated MPXV. The SERS and photothermal LODs for inactivated MPXV were 102 copies/mL. The platform also showed high specificity against other prevalent respiratory viruses and maintained xcellent inter-batch reproducibility across all detection modalities. These results indicat that the MoS2@Au-based M-ICA platform is a highly sensitive and accurate diagnostic tool for MPXV detection, offering a promising approach for immunodiagnosis in diverse application scenarios.
