Antimicrobial resistance (AMR) necessitates rapid and reliable diagnostic technologies to support timely antimicrobial selection and stewardship. Although conventional culture-based antimicrobial susceptibility testing (AST) remains clinically important, its extended turnaround time limits timely intervention, while molecular diagnostics can rapidly identify resistance determinants but may not accurately predict phenotypic resistance. This review critically examines the integration of artificial intelligence (AI) with biomedical microdevices and biosensors for rapid AMR detection, encompassing microfluidic and lab-on-chip platforms, electrochemical and optical biosensors, and CRISPR-based technologies. Emphasis is placed on the application of AI for signal processing, noise reduction, pattern recognition, multiplexed detection, and predictive interpretation of complex biosensor outputs. Key translational challenges, including biofouling, matrix interference, sample-preparation variability, bubble formation, evaporation, signal drift, invalid-test rates, device reproducibility, and inadequate standardisation of reporting and validation, are critically evaluated. The requirements for embedded AI, including memory, computational resources, latency, and energy efficiency, are also considered for portable point-of-care deployment. Analytical, clinical, and operational validation are distinguished, with emphasis on prospective multicentre evaluation using representative specimens and reference AST methods. Beyond clinical diagnosis, AI-enabled microdevices offer potential for decentralised AMR surveillance across wastewater, agricultural, veterinary, and environmental settings within a One Health framework. Translation will require standardised datasets, generalisable and explainable AI, scalable manufacturing, regulatory oversight, data governance, and equitable deployment. Overall, AI-integrated biomedical microdevices represent a promising approach to rapid, decentralised, and actionable AMR diagnostics and surveillance, contingent on robust real-world validation.
Neurodegenerative diseases, including Alzheimer’s disease (AD) and Parkinson’s disease (PD), represent a growing health burden, while early diagnosis remains challenging because established diagnostic approaches, including cerebrospinal fluid analysis and neuroimaging, are invasive, expensive, and unsuitable for decentralized point-of-care testing. Recent progress in electrochemical and paper-based biosensors has enabled low-cost, portable, and multiplexed biomarker analysis. However, their clinical translation remains limited by low biomarker concentrations, matrix interference, nonspecific adsorption, sensor-to-sensor variability, storage stability, and the need for validation using clinical samples. Herein, a paper-based multiplex electrochemical aptasensor was developed for simultaneous impedimetric detection of amyloid beta-42 (Aβ-42) and alpha synuclein-60 (αS-60), biomarkers associated with AD and PD, respectively. The device comprised screen-printed carbon and silver electrodes on a cellulose substrate, two independently functionalized working electrodes, zinc oxide nanorods (ZnO-NRs) as electroactive signal-amplification layers, and target-specific DNA aptamers as biorecognition elements. Electrochemical impedance spectroscopy was used for label-free measurement of target-induced changes in charge-transfer resistance. The sensor exhibited a linear response from 0.1 ng/mL to 10 µg/mL for both biomarkers, with limits of detection of 4.41 ng/mL for Aβ-42 and 2.62 ng/mL for αS-60. The platform further detected both analytes in artificial cerebrospinal fluid and artificial saliva, demonstrating that the paper-based sensing interface remains responsive in physiologically mimicked matrices. Overall, this study provides a low-cost, disposable, label-free, and multiplexed proof-of-concept platform for simultaneous AD- and PD-associated biomarker detection. Although the system demonstrates promising performance in artificial matrices, further optimization of sensitivity, surface immobilization, antifouling behaviour, long-term storage, and validation using real clinical saliva and CSF samples will be required before practical diagnostic deployment.
Circulating tumor cells (CTCs), shed from primary tumors into the peripheral bloodstream, are important biomarkers for early cancer diagnosis and disease monitoring. Deterministic lateral displacement (DLD) microfluidic devices enable label-free CTC isolation through continuous, size-based separation. However, DLD performance is highly sensitive to coupled geometric and operating parameters, making optimal design challenging. Conventional trial-and-error approaches are computationally expensive and provide limited insight into parameter interactions, necessitating data-driven optimization. Here, a Bayesian optimization (BO) framework integrating high-fidelity numerical simulations with Gaussian process (GP) surrogate modeling is developed for efficient DLD design. Using pillar gap size, post-array tilt parameter and flow rate as design variables, BO iteratively constructs a surrogate of the nonlinear target-particle lateral-displacement response and guides adaptive sampling. The GP model achieved high predictive accuracy, with an R^2 of 0.9873 and a mean absolute error of 1.24μ m on an independent validation dataset. Compared with Evolution Strategy (ES), BO reduced the evaluations required to identify the best-performing design within the prescribed design space by 67.2 75.51μ m . Post-optimization particle-size and outlet-region analyses identified an effective critical diameter of approximately 11.4μ m and demonstrated distinct collection of the investigated NSCLC CTC target surrogate and smaller blood-cell-relevant background-size particles across three release positions. The GP surrogate also enables response-surface mapping, global sensitivity analysis and uncertainty quantification. Results identify the post-array tilt parameter as the dominant factor governing target-particle lateral displacement, while flow rate strongly interacts with geometric variables. Uncertainty analysis confirms stable performance under perturbations. The proposed framework provides an efficient, interpretable, and reliability-aware approach for DLD microfluidic device design.
Nanomedicine has progressed far beyond its early role as an experimental drug-carrier toolbox and today stands as a clinically validated enabling technology. Liposomal formulations and albumin-bound nanoparticles have transformed cancer therapy, while lipid nanoparticle (LNP) platforms accelerated the rapid development and global deployment of SARS-CoV-2 mRNA vaccines—demonstrating how nanoscale engineering can reshape therapeutic response, manufacturing speed, and public health impact. With these successes as foundation, the field is rapidly expanding into more sophisticated delivery systems including cell-hitchhiking nanoparticles, biomimetic multivalent vaccines, implantable immunotherapy depots, high-loading polymeric micelles, organ-targeted gene and RNA carriers, and bio-hybrid extracellular vesicle and mitochondrial therapies. These platforms aim not only to transport drugs, but to modulate immunity, direct regeneration, and enable precision intervention at the cellular and molecular scale. This perspective summarizes the current landscape of biomedical nanotechnologies and outlines how their continued evolution positions nanomedicine as an enabling science driving the next generation of therapeutics.
Microfluidic paper-based analytical devices (µPADs) have gained attention for clinical diagnostics due to low cost, simplicity, and suitability for point-of-care testing. However, high reagent and sample consumption, limited analytical efficiency, and non-optimized designs still restrict broader use. To overcome these limitations, we propose a miniaturized and re-optimized µPAD design for colorimetric dopamine that reduces the required assay time and minimizes reagent and sample consumption while maintaining reproducible analytical performance in biological samples. To achieve this goal, six distinct µPAD geometries were designed and evaluated to examine the effect of device architecture on colorimetric signal development, followed by optimization of the selected geometry and reaction conditions. Based on its temporal color development profile and signal intensity, the most suitable configuration was selected for further optimization This optimized platform was further evaluated or reagent consumption, response time, reproducibility, and analytical performance in biological samples. The optimized miniaturized µPAD design achieved more than a six-fold reduction in reagent usage and a two-minute decrease in drying time. The optimized configuration showed an LOQ of 0.0517 µmol/L in buffer, while the selectivity experiments showed signal deviations within ± 3
