Search Results (1,512)

With the rapid deployment of deep neural networks (DNNs) on edge devices, traditional hardware accelerators face significant challenges in terms of data security, computational redundancy caused by sparsity, and uneven utilization of on-chip resources. This paper proposes SaE-FPGA, a secure and efficient DNN accelerator designed specifically for edge FPGA platforms. The architecture introduces three core innovations: (1) Hash-Bypass Processing Unit (HBPU): Integrating a high-speed SHA-256 hardware engine with a hash-sparse bitmap mechanism, it enables real-time data integrity verification within a single clock cycle while skipping computations for redundant zero-value data. (2) Flexible Mixed-Precision Processing Element (FMP): By reconfiguring idle BRAM and LUT resources into an active lookup table multiplication engine, it overcomes the physical bit-width limitations of DSP blocks and supports INT8/INT6/INT4 mixed-precision multiplication. (3) Multi-mode Reconfigurable Streaming Frame (MRSF): A sparse-aware, elastic load balancing and data routing mechanism designed to mask long memory access latencies and ensure high hardware resource utilization. Experimental results on the Zynq 7045 platform demonstrate that SaE-FPGA reduces redundant computations by 23.2% while maintaining high precision and minimizing precision loss. The system effectively mitigates the risk of physical tampering. When tested on ResNet-50, it achieved a 27.2% improvement in energy efficiency and a 2.97× speedup compared to DSP-based FPGA solutions. Furthermore, by fully exploiting the hybrid BRAM-LUT and DSP configuration, the proposed accelerator achieves a remarkable peak throughput of 782.4 GOPS. Full article

Implementing cryptographic primitives on resource-constrained IoT devices involves tight latency, code-size, and energy budgets. This work proposes a general LLVM backend instruction-selection strategy that recognizes single-bit update idioms—typically expressed as LOAD–-(AND/OR)–-STORE sequences in SHA-256 and similar bit-oriented code—and lowers them to the most efficient target-specific bit-manipulation primitive when legality and cost conditions are met. As a concrete instantiation, we implement the strategy for the Renesas RL78/G23 ISA by rewriting eligible patterns into SET1/CLR1 instructions when the constant mask targets exactly one bit. We evaluate the resulting backend on an RL78/G23 platform using cycle counts and code size (bytes) across SHA-256-driven workloads motivated by firmware integrity checking, Merkle-tree hashing, HMAC-based authentication, password-based key derivation (PBKDF2), and chunk-level update validation. The observed cycle reductions are also converted to absolute time across the device’s supported on-chip oscillator frequencies to quantify latency impact under different clocking modes. The experimental validation in this work is limited to the RL78/G23 backend implementation. The underlying instruction-selection idea may be adaptable to other RL78-family devices or to other embedded architectures that provide equivalent single-bit set/clear or bitfield operations; however, such adaptations require target-specific legality checks, cost modeling, and separate experimental validation. Full article

Patient-derived organoids (PDOs) have rapidly transitioned from research tools into promising platforms for clinical translation. In this review, we analyze 139 PDO-related clinical trials registered between 2023 and 2025 and contrast them with recent advances in disease modelling. Our analysis revealed a predominance of oncology-focused studies, with translational maturity spanning from foundational research to studies in which PDOs directly informed clinical decision-making. In contrast, non-oncology areas show extensive preclinical progress but remain trial-poor. We found that trial registration is geographically concentrated in a small number of countries, reflecting uneven global adoption. We then explored advances in disease modeling, mainly confined to preclinical studies, including immune-competent PDOs, complex organ-on-a-chip systems, synthetic matrices, AI-enabled platforms, and therapeutic transplantation. Based on these findings, we propose a conceptual framework outlining the trajectory of PDO adoption in clinical trials. This trajectory can be understood as three overlapping waves of translation: the first wave, focusing on oncology, has already demonstrated impacts on patient care; the second, targeting non-oncology diseases, is scientifically advanced but has not achieved widespread clinical application; and the third, involving frontier technologies, remains in the preclinical stage. Understanding these trajectories underscores the promise and challenges of PDOs that must be addressed for broader clinical adoption. Full article

The picking of tea leaves in tea gardens requires multiple batches in the short and valuable tea harvest period. To realize timely and efficient tea plucking, it is feasible to use unmanned aerial vehicles (UAV) for tea shoot detection in large tea gardens. For the typical small targets of tea buds in unmanned aerial vehicle (UAV) aerial images, it is necessary to design an efficient tea buds detection model. In order to improve the accuracy and the speed of the tea buds detection in the UAV images, we designed the SH-CoordMapping hash space mapping algorithm to accelerate the remerging of the detection results into the original image. The C2PSA-BI module and the CARAFE upsampling module are applied to improve detail preservation during feature fusion. A lightweight detection head is further used to reduce redundant computation in the detection stage. By comparing with the traditional detection methods, it can be proved that the SWO sections are necessary for UAV-scale tea shoots detection. Based on the accuracy and the number of model parameters, the YOLO11n model with slice size as 640 and overlap rate as 0.1 performs the best. The TSDet-UAV was deployed on the NVIDIA Jetson Orin NX chip to construct an inspection system capable of real-time acquisition and detection. The experimental results demonstrate that the proposed TSDet-UAV achieves excellent performance, recording an mAP₅₀ of 52.9% on the constructed UAV-TS dataset while maintaining high efficiency. With a parameter size of 2.4 M and a total processing time of 1.32 s per high-resolution image under TensorRT FP16, the processing speed is highly suitable for real-time edge deployment on agricultural UAV platforms. The UAV image-based tea garden shoot inspection platform proposed in this paper can effectively confirm the growth status of tea shoots, assisting farm management in formulating precise picking plans. Full article

Many System-on-Chip (SoC) studies rely solely on simulation and tool-based results, encountering unexpected failures during post-silicon validation. In particular, silicon-level demonstrations of Hardware/Software (HW/SW) functional equivalence, which confirms that an FPGA-validated design operates identically on an ASIC with the same firmware, remain extremely rare. This work proposes a reproducible FPGA-to-silicon verification methodology that establishes HW/SW functional equivalence at the silicon level by applying an identical firmware source code, device driver, and memory map to both platforms. The methodology is validated on an Arm Cortex-M0-based SoC platform fabricated in Samsung 28 nm Low Power Plus (LPP) CMOS technology with a dual Inter-Integrated Circuit (I2C) interface. The fabricated chip integrates two 64KB on-chip memories within a core area of 653 $\mathsf{\mu }$m × 769 $\mathsf{\mu }$m, operates at 125 MHz, and consumes 17.5 mW at the optimal operating point of 1.0 V. The primary contributions are: (1) a reproducible FPGA-to-silicon HW/SW functional equivalence verification methodology based on shared firmware source code, device driver, and memory map across both platforms, (2) silicon-measurement-based performance characterization with verified experimental data, (3) a reproducible design methodology documenting the complete flow from FPGA verification through ASIC fabrication, including static timing closure, place-and-route, and physical verification, and (4) an extensible SoC platform architecture enabling researchers to integrate and validate their own Intellectual Property (IP) via Advanced High-performance Bus (AHB) and I2C interfaces. Full article

This work presents the Smart-Reflex-Scope, a compact and accessible smartphone-based microscope with integrated reflective illumination developed for on-chip analysis of microparticle dynamics. In this work, the platform is specifically employed to characterize size-dependent microparticle motion within a microchannel. The Smart-Reflex-Scope simultaneously functions as an illumination source and imaging unit by integrating a reversed smartphone camera lens, a custom reflex module, a microfluidic chip, and a precision Z-axis translation stage for focal adjustment. The optical performance was quantitatively evaluated in terms of equivalent focal length, magnification, and object-plane spatial resolution, providing a comprehensive assessment of the system’s microscale imaging capabilities. A comparative design study was conducted between two configurations: Design-1, based on normal reflection, and Design-2, based on angular reflection. The two approaches were analyzed with respect to illumination uniformity and imaging performance to identify the optimal configuration for enhanced visualization. Experimental validation was performed using synthetic microparticles with diameters of $6\mathsf{\mu }\mathrm{m}$ and $20\mathsf{\mu }\mathrm{m}$, enabling assessment of the system’s ability to resolve and dynamically track micrometric objects of different sizes. The results demonstrate reliable detection and size-dependent dynamic characterization. A two-factor statistical ANOVA analysis confirmed the significance of the observed differences between microparticle groups under the tested experimental conditions (p-value $<0.0001$). Overall, the proposed platform represents a scalable and miniaturized microscopy solution bridging conventional benchtop instruments and portable analytical devices. Full article

Neurological disorders, diabetes, cancer, and infectious diseases remain major global health concerns, particularly in low- and middle-income countries with insufficient access to accurate and rapid diagnostics. Conventional biochip sensing platforms, while effective, are often constrained by complex instrumentation and have limited capability for handling complex and large datasets. This review aims to address these limitations by evaluating the integration of artificial intelligence (AI) with biochip technology improve biomedical diagnostics. We systematically analyze recent advances in AI-integrated biochips, such as spectroscopic, paper-based, lab-on-chip, and microfluidic platforms integrated with reinforcement learning, machine learning, and deep learning models. These pre-trained AI models simplify pattern recognition, feature extraction, and automated data processing from a variety of biosensor outputs, such as electrochemical, fluorescence, and colorimetric signals. The reviewed studies indicate improved real-time diagnostic sensitivity and accuracy across biomedical applications. Overall, we discuss ongoing challenges and future perspectives toward explainable, robust, and smartphone-assisted AI-integrated biochips for rapid and accurate diagnostics. The review offers a comprehensive overview of AI-integrated biochips to support effective disease detection and clinical decision-making. Full article

Cell-based immunotherapies require noninvasive tools that can quantify the migration, biodistribution, and persistence of administered immune cells. This review focuses primarily on oncologic immune cell therapy, while also considering selected inflammatory disease models in which immune-cell trafficking is biologically relevant. We critically compare direct radionuclide labeling, sodium iodide symporter (NIS)-based reporter gene imaging, radionuclide-integrated nanoplatforms, and Cerenkov-based hybrid optical conversion strategies. Direct labeling with agents such as [⁸⁹Zr]Zr-oxine, [¹¹¹In]In-oxine, and [⁹⁹ᵐTc]Tc-HMPAO enables early positron emission tomography (PET)/single-photon emission computed tomography (SPECT) biodistribution assessment, usually within hours to several days after cell administration. NIS reporter imaging with [¹²⁴I]NaI, [¹²³I]NaI, [⁹⁹ᵐTc]TcO₄⁻, or [¹⁸F]TFB supports repeated viability-dependent imaging, because signal generation depends on active transporter expression in living engineered cells. Radionuclide-integrated gold nanoplatforms can improve intracellular retention and offer theranostic potential through combined imaging, photothermal, radiotherapeutic, or immunomodulatory functions. We further discuss PET/SPECT balance, radiopharmaceutical nomenclature, nanoparticle stabilization, ethical aspects of genetic modification, tumor-on-a-chip systems for preclinical testing, and limitations of narrative evidence synthesis. Together, these platforms provide complementary strategies for image-guided immune cell therapy, with translational relevance for patient selection, treatment optimization, safety monitoring, and oncology practice. In conclusion, NIS-centered nuclear imaging and radionuclide-integrated nanoplatforms represent complementary, clinically actionable tools for quantitative immune-cell tracking, therapeutic optimization, and safety monitoring in translational oncology and inflammatory disease research. Full article

Glioblastoma multiforme (GBM) remains the most aggressive and therapeutically intractable primary brain tumor, with many patients experiencing rapid relapse despite maximal surgical resection followed by standard chemoradiation. This persistent failure reflects the convergence of profound tumor-intrinsic genetic heterogeneity and a highly dynamic, spatially structured, and immunosuppressive tumor microenvironment (TME). Together, these forces create strong selective pressures that fuel tumor evolution, intratumoral diversity, phenotype plasticity, diffuse invasion, and robust resistance to therapy. The TME of GBM is orchestrated through a complex interplay between diverse cellular constituents, including tumor-associated macrophages, reactive astrocytes, endothelial cells, pericytes, and GBM stem cells, and non-cellular components such as extracellular matrix remodeling, hypoxia, metabolic and nutrient gradients, and spatially patterned cytokine and chemokine signaling networks. Additionally, heterogeneity in blood–brain barrier (BBB) and blood–tumor barrier (BTB) complicates drug delivery and immune surveillance, reinforcing therapeutic resistance and regional tumor adaptation. Conventional two-dimensional cell cultures and animal models fail to sufficiently capture these multiscale, patient-specific interactions, limiting their translational predictive power. In this narrative review, we synthesize recent advances in GBM organoid technologies as physiologically relevant, three-dimensional platforms that more faithfully recapitulate TME for driving tumor evolution and treatment resistance. We compare complementary organoid strategies, including patient-derived GBM organoids that preserve native cytoarchitecture, cerebral organoid co-culture systems that reconstruct tumor–brain interactions, and advanced platforms incorporating immune and vascular features such as air–liquid interface cultures, microglia-enriched systems, and BBB/BTB-integrated models. Finally, we highlight emerging innovations such as spatial transcriptomics, organoid-on-a-chip systems, live imaging coupled with lineage tracing, genome engineering, and artificial intelligence integration that collectively position GBM organoids at the forefront of precision neuro-oncology, reproducing TME, enabling dynamic mapping of tumor evolution, and accelerating patient-specific therapeutic discovery. Full article

Background: Internode length (IL), a key component of plant height (PH), plays an important role in achieving the optimal architecture in wheat. However, the genetic mechanisms underlying internode elongation are not well understood. Methods: In this study, a recombinant inbred line (RIL) population derived from a cross between Bainong 4199 (BN4199) and Zhengyinmai 2 (ZYM2) was evaluated for PH and five ILs across two field locations over two years and genotyped using a 120 K liquid-phase chip. Results: A total of 141 quantitative trait loci (QTL) associated with PH and the five ILs were mapped onto 20 chromosomes, except for chromosome 5D. Among these, 37 stable QTL were identified on chromosomes 1B, 2B, 2D, 4B, 5A, 7A, 7B and 7D, accounting for 3.86–25.97% of the phenotypic variation. Meanwhile, 23 co-localized QTL associated with at least two traits were detected, with QTL cluster regions on chromosomes 2D, 4B, 5A, 7A, and 7B. Moreover, the total additive effects of the QTL combinations increased with the number of QTL, which indicates the effectiveness of pyramid breeding. Additionally, based on gene function annotation, the cloning and characterization of rice orthologs, and analysis via the QTG miner module of the wheat integrative gene regulatory network (wGRN) platform, 63 candidate genes (e.g., Rht1, Rht8, TB1 and ZnF-B) were prioritized within the stable QTL intervals, and their tissue expression patterns were analyzed. Conclusions: Collectively, these findings not only deepen our understanding of the genetic basis of PH and ILs in wheat but also lay a foundation for the further validation and functional characterization of candidate genes, enabling the optimization of plant architecture through marker-assisted selection (MAS) to ultimately improve agronomic performance and yield potential. Full article

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a widespread condition with a complex pathogenesis. Cell-based models are important tools for studying the mechanisms underlying its development and progression. The aim of this review is to analyze the HepaRG, Huh-7, immortalized human hepatocyte (IHH), and primary human hepatocyte (PHH) cell lines for modeling and studying MASLD. HepaRG represents the most metabolically competent immortalized hepatocyte model with preserved biotransformation activity and a physiological bioenergetic response to lipid loading, making it valuable for pharmacological and toxicological studies. Huh-7 is distinguished by its accessibility and suitability for studying steatosis, lipotoxicity, insulin resistance, and paracrine mechanisms of fibrogenesis; however, its use is limited by its tumor origin, impaired carbohydrate metabolism, and low activity of xenobiotic-metabolizing enzymes. The IHH model occupies an intermediate position because of its non-tumor origin and is of interest for studies of senescence, epigenetic regulation, and signaling pathways involved in steatosis, although interpretation of results requires consideration of immortalization-related effects and specific metabolic limitations. PHH remains the most physiologically relevant platform for MASLD modeling, particularly in three-dimensional (3D) and microphysiological formats; however, its use is limited by high cost, interindividual variability, and the limited duration of the differentiated phenotype. Increasing model complexity—from two-dimensional (2D) monocultures to co-cultures, spheroids, and organ-on-chip systems—enhances physiological relevance and enables reproduction not only of steatosis but also of the inflammatory and fibrogenic components of MASLD progression, yet it reduces reproducibility and complicates standardization. Overall, none of the existing models is universal, and the optimal strategy is to select models according to the specific research question. A key direction for future research is the standardization of steatosis induction protocols and the unification of criteria for evaluating results. Full article

Series arc faults on the DC side of photovoltaic (PV) systems are a critical hazard that can trigger system fires. Conventional contact-based detection methods suffer from cumbersome installation and high retrofit cost, whereas existing non-contact approaches mostly rely on megahertz-level high-frequency sampling and therefore require expensive radio-frequency instrumentation or high-performance computing platforms. As a result, it remains difficult to simultaneously achieve strong interference immunity and real-time performance on low-cost embedded devices with limited resources. To address this engineering paradox between high-frequency sampling and constrained computational capability, this paper proposes a fully embedded, non-contact arc fault detection system based on a 12–80 kHz low-frequency sub-band selection strategy. By exploiting the physical characteristic of broadband energy elevation induced by arc faults, the proposed strategy avoids dependence on high-bandwidth hardware. Guided by this strategy, a Moebius-topology coaxial shielded loop antenna is employed as the near-field sensor, while an ultra-simplified passive analog front end is constructed directly by using the on-chip programmable gain amplifier and analog-to-digital converter of the microcontroller unit, enabling efficient signal acquisition and fast Fourier transform processing within the target sub-band. To cope with complex background noise in the low-frequency range, an environment-adaptive baseline mechanism based on exponential moving average and exponential absolute deviation is developed for dynamic decoupling. In addition, a lightweight INT8-quantized multilayer perceptron is introduced as a nonlinear auxiliary module, thereby forming a robust hybrid decision architecture with complementary rule-based and artificial intelligence components. Experimental results show that, under the tested household, laboratory, and PV-site conditions, the proposed system achieved an overall detection rate of 97%, while the remaining 3% mainly corresponded to failed ignition or non-sustained arc attempts rather than persistent false triggering during normal monitoring. Full article

We present a modular microfluidic platform for constructing miniaturized, compartmentalized cell culture systems that support monoculture, co-culture, and organ-on-a-chip models of human tissues. The devices provide architecturally defined three-dimensional microenvironments in which heterogeneous cell populations can be cultured in close proximity while maintaining precise spatial organization and independent access to each compartment. In vivo-like perfusion into, from, and between adjacent chambers is achieved via micro-engineered porous barriers that act as perfusion microchannels, enabling controlled convective and diffusive transport and recapitulating paracrine signaling between tissue units. As a proof of concept, we implement an adipose–immune co-culture model that reproduces key features of inflamed, insulin-resistant adipose tissue, including altered cytokine secretion and glucose uptake. Together, these features establish a versatile platform for the biofabrication of customizable single-organ and multi-organ in vitro models that more faithfully recapitulate human tissue structure and function for applications in disease modeling, immunometabolic studies, and preclinical drug testing. Full article

Soil eco-enzymes (i.e., microbial extracellular enzymes) play essential roles in terrestrial nutrient cycling and support ecosystem services. In this regard, their activities serve as indicators of soil health. However, conventional spectrophotometric and microplate fluorometric assays are often limited by lengthy reaction procedures, relatively high reagent consumption, and insufficient compatibility with complex soil matrices. In this investigation, we developed a portable, centrifugally driven microfluidic chip for the rapid and sensitive determination of multiple soil extracellular enzyme activities. This integrated platform automated sample aliquoting, reagent metering, mixing, and sedimentation, enabling the parallel measurement of eight enzymes. Such system demonstrated precise liquid control via capillary valves and high optical uniformity (<5% fluorescence variation). 4-methylumbelliferone (MUF)-based calibration exhibited strong linearity (R² > 0.99) across diverse soil types. Compared with conventional microplate assays, the microfluidic method improved reproducibility (CV < 15%), enhanced the detection of weak fluorescence signals, and increased throughput while reducing reagent consumption. This field-ready platform provides a robust solution for standardized soil enzyme assessment and offers future potential for integration with AI-driven data analytics and large-scale ecological monitoring frameworks. Full article

Invasive fungal respiratory infections (IFRIs) remain a major cause of morbidity and mortality among immunocompromised patients, yet diagnosis continues to be hindered by nonspecific clinical features, limited sample accessibility, and the poor sensitivity or specificity of conventional tests. Microfluidic and microelectromechanical systems (MEMS)-based biosensing platforms have emerged as promising alternatives, enabling rapid, minimally invasive, and highly specific detection of fungal pathogens and host responses. Microfluidic nucleic acid and antigen assays allow on-chip amplification and immunodetection with reduced sample volumes and turnaround times, while CRISPR-enhanced systems further improve analytical sensitivity. Parallel advances in host response profiling—including transcriptomic, proteomic, and cytokine-based signatures—have demonstrated feasibility for integration into lab-on-a-chip platforms. MEMS-based technologies extend this potential by facilitating real-time analysis of exhaled volatile organic compounds, mechanical biosensing of fungal DNA and antigens, and in situ monitoring of device-associated biofilms. Translational studies highlight potential applications across intensive care, hematology–oncology, and transplant settings, as well as in outpatient monitoring of high-risk populations. However, several challenges remain, including limited multicenter validation, matrix-related biofouling effects, and a lack of standardization in fungal biomarker panels. Future directions include AI-driven interpretation of multianalyte data, multiplexed integration of host and pathogen markers, and development of fully cartridge-based systems for near-patient deployment. Collectively, these innovations may shift fungal diagnostics toward earlier, more precise, and patient-tailored interventions, improving outcomes in vulnerable populations. Full article