Advanced Sensor Research最新文献

Biofluid Sensors

The cover illustrates the selective sensing of a salivary biomarker for oral diseases — sialic acid, using an electrochemical molecularly imprinted polymer sensor. The authors introduce a disposable electrochemical sensor based on a molecularly imprinted polymer of aminophenylboronic acid electropolymerized directly on laser-induced graphene electrode, enabling selective detection of sialic acid. More details can be found in the Research Article by Alexander V. Shokurov, Max Nobre Supelnic, and Carlo Menon (DOI: 10.1002/adsr.202500156).

Fluorescent sensor arrays provide pattern‑based, multidimensional optical fingerprints for detecting chemically and biologically diverse analytes across complex matrices. By leveraging orthogonal readouts (intensity, ratiometric channels, lifetime, and excitation–emission matrices) from cross-reactive and target-specific elements, fluorescent sensor arrays achieve sensitive, rapid measurements suitable for environmental, biomedical, and food‑safety applications. The data richness of fluorescent sensor arrays, however, exceeds the capabilities of traditional analytical approaches. Classical chemometrics, exemplified by principal component analysis for exploratory visualisation and linear discriminant analysis for baseline classification, assumes linear structure and homoscedasticity, and therefore struggles with non‑linear photophysical responses, multicollinearity, and mixture quantification. This review surveys machine‑learning methods that address these limitations for both discrimination and quantification, including support vector machines and k‑nearest neighbours, tree ensembles, Gaussian process and support‑vector regression, and neural/deep‑learning models tailored for spectra, excitation–emission matrices, and images. Practical guidance is provided on acquisition and pre‑processing, rigorous validation (nested cross‑validation, external tests), uncertainty quantification, and interpretability to inform array design and deployment. Case studies demonstrate improved sensitivity, selectivity, robustness, and calibration transfer. Remaining challenges, dataset size, drift, and matrix effects, are discussed alongside opportunities in excitation‑multiplexed “virtual arrays”, active learning, and explainable AI for next‑generation, data‑driven fluorescent sensing.

Fluorescent sensor arrays provide pattern‑based, multidimensional optical fingerprints for detecting chemically and biologically diverse analytes across complex matrices. By leveraging orthogonal readouts (intensity, ratiometric channels, lifetime, and excitation–emission matrices) from cross-reactive and target-specific elements, fluorescent sensor arrays achieve sensitive, rapid measurements suitable for environmental, biomedical, and food‑safety applications. The data richness of fluorescent sensor arrays, however, exceeds the capabilities of traditional analytical approaches. Classical chemometrics, exemplified by principal component analysis for exploratory visualisation and linear discriminant analysis for baseline classification, assumes linear structure and homoscedasticity, and therefore struggles with non‑linear photophysical responses, multicollinearity, and mixture quantification. This review surveys machine‑learning methods that address these limitations for both discrimination and quantification, including support vector machines and k‑nearest neighbours, tree ensembles, Gaussian process and support‑vector regression, and neural/deep‑learning models tailored for spectra, excitation–emission matrices, and images. Practical guidance is provided on acquisition and pre‑processing, rigorous validation (nested cross‑validation, external tests), uncertainty quantification, and interpretability to inform array design and deployment. Case studies demonstrate improved sensitivity, selectivity, robustness, and calibration transfer. Remaining challenges, dataset size, drift, and matrix effects, are discussed alongside opportunities in excitation‑multiplexed “virtual arrays”, active learning, and explainable AI for next‑generation, data‑driven fluorescent sensing.

Accurate intraoperative identification of brain tumor margins remains a major challenge in neurosurgery. Tumors often differ from healthy brain tissue in their mechanical properties, such as stiffness and viscoelasticity, yet current imaging methods provide limited real-time mechanical feedback during surgery. In this study, the use of acoustic sensing based on surface acoustic wave (SAW) actuators to distinguish between non-neoplastic brain tissue, primary brain tumors, and metastatic tumors based on their acoustic properties is investigated. Tissue samples are measured ex vivo, and attenuation is analyzed as a function of mass and stiffness. Results showed clear, consistent trends, where non-neoplastic tissues exhibit increased acoustic attenuation, metastatic tumors exhibited intermediate attenuation, and primary tumors showed the lowest attenuation, reflecting increasing stiffness across these tissue types. These findings align with previously reported mechanical properties from techniques such as magnetic resonance elastography and microindentation, where acoustic/SAW based methodologies have significant potential advantages in throughput, cost-effectiveness and integrability with other techniques. Accordingly, this work demonstrates that SAW sensing enables reliable sensitivity to biomechanical differences between tissue types, supporting its potential as a real-time, non-invasive tool for intraoperative tumor detection.

Electrochemical Immunosensor

Glycoprotein 88 (GP88) is a secreted biomarker that is overexpressed in various cancers, as well as in neurological and inflammatory diseases. In the Research Article (DOI: 10.1002/adsr.202500122), Anna Ignaszak and co-workers introduce the first disposable electrochemical immunosensor built on a screen-printed electrode for the detection of a circulating GP88.

Research on chloride sensors is important for applications in the food industry, drinking water quality control, and medicine for the determination of cystic fibrosis (CF). In this work, we report the chemical synthesis of silver nanoparticles (AgNPs) used as chloride sensors. These AgNPs were characterized by XRD, TEM, and electrochemical techniques. The chloride sensor response showed a linear range between 0 and 160 mm chloride in PBS (pH 5.6) and a sensitivity of 55.4 ± 0.3 mV/decade. In addition, a MOSFET transistor was developed as a QA voltage source for coupling the transducer to amplify the signal by 1 610.27% of the initial response, reducing quantification times to less than 40 s, analyzing results, and transmitting data to a mobile device via WI-FI. A linear range of 0 to 160 mm and a sensitivity of 0.012 V mm⁻¹ with high reproducibility were achieved with the MOSFET transistor. This chloride sensor enables fast and efficient measurement of chlorides using AgNPs and a MOSFET transistor, which could be used for monitoring and auxiliary diagnosis of cystic fibrosis.

Advancing drug development and disease modeling requires physiologically relevant in vitro systems that accurately reproduce the dynamic and selective transport functions of human tissue barriers. In response to regulatory efforts to reduce animal testing, there is increasing demand for standardized, scalable, and sensor-integrated microphysiological platforms. Although organ-on-chip technologies have improved biological relevance, many remain limited by technical complexity and insufficient sensing, particularly for non-equilibrium transport processes that characterize barriers such as the renal tubule, intestine, bloodbrain barrier, and placenta. Here, we present a fully automated microfluidic platform for dynamic and quantitative characterization of molecular transport across biomimetic interfaces. The system emulates key biophysical features, including directional flow, controlled gradient formation, and tunable shear stress, using synthetic, cell-free architectures. Integrated oxygen sensors enable monitoring of physiologically relevant oxygen levels, while embedded flow sensors support closedyloop control of fluid dynamics and solute delivery. Together, these sensing capabilities allow experimental control and time-resolved analysis of molecular transport under non-equilibrium conditions. Validated using acellular models, the platform is engineered for future integration with living tissues and molecular sensorization. Proof-of-concept studies demonstrate its ability to reproduce key transport dynamics, highlighting its potential for pharmacokinetic modeling, organ-on-chip research, and preclinical drug development.

This project explores the development and feasibility of a non-invasive biosensor system designed to capture cardiac signals and evaluate the validity of estimating blood pressure over a 24-hour period using a single-lead electrocardiogram (ECG). The system integrates an Apple Watch, a laboratory-built digital stethoscope, and a consumer-grade blood pressure cuff to collect cardiac signals that are processed through a cloud-based pipeline. An artificial intelligence (AI) model was trained on the publicly available Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II) clinical dataset, which generates a 24-hour continuous blood pressure prediction waveform from the ECG data, and Autonomic Aging ambulatory dataset. This predicted waveform is averaged and compared with the cuff-based blood pressure readings to assess accuracy and signal reliability. To further validate the system, cross-device agreement in beats per minute (BPM) is used to demonstrate the plausibility of cuffless blood pressure tracking. Overall, this study presents a proof-of-concept framework for continuous, cuffless blood-pressure estimation using ECG alone. By combining mechanical heartbeat verification with cloud-based AI processing, the system demonstrates technical feasibility and provides a foundation for non-invasive, long-term monitoring and future clinical validation.

This work aimed to create a capacitive force sensing system based on flexible materials and structures, miniaturizing the hardware by 87% and the interface, and improving the stability of the signal processing module (from force to capacitance) for robot applications in a wearable shape. The system includes a force sensor, signal acquisition integrated circuit (IC), microcontroller unit, Bluetooth IC, and lithium-polymer battery. The sensor was composed of polymeric materials and elastomers, which were connected to the wristwatch-shaped transmission port before the signal was wirelessly analyze in real-time. Field tests indicated that the responses exhibited reasonable tolerance (less than 2%), reliable short- and long-term stability (variation less than 5%), and remarkable repeatability (linear coefficient of determination of 0.9975). The wristwatch-shaped transmission port thus demonstrated its superiority in practical application and exhibited novelty from the viewpoint of modularization, with balanced characteristics among similar solutions, which provided add-on functions to existing robots.