Innovative Architecture Research Articles

CTCNet: a fine-grained classification network for fluorescence images of circulating tumor cells.

The identification and categorization of circulating tumor cells (CTCs) in peripheral blood are imperative for advancing cancer diagnostics and prognostics. The intricacy of various CTCs subtypes, coupled with the difficulty in developing exhaustive datasets, has impeded progress in this specialized domain. To date, no methods have been dedicated exclusively to overcoming the classification challenges of CTCs. To address this deficit, we have developed CTCDet, a large-scale dataset meticulously annotated based on the distinctive pathological characteristics of CTCs, aimed at advancing the application of deep learning techniques in oncological research. Furthermore, we introduce CTCNet, an innovative hybrid architecture that merges the capabilities of CNNs and Transformers to achieve precise classification of CTCs. This architecture features the Parallel Token mixer, which integrates local window self-attention with large-kernel depthwise convolution, enhancing the network's ability to model intricate channel and spatial relationships. Additionally, the Deformable Large Kernel Attention (DLKAttention) module leverages deformable convolution and large-kernel operations to adeptly delineate the nuanced features of CTCs, substantially boosting classification efficacy. Comprehensive evaluations on the CTCDet dataset validate the superior performance of CTCNet, confirming its ability to outperform other general methods in accurate cell classification. Moreover, the generalizability of CTCNet has been established across various datasets, establishing its robustness and applicability. What is more, our proposed method can lead to clinical applications and provide some help in assisting cancer diagnosis and treatment. Code and Data are available at https://github.com/JasonWu404/CTCs_Classification .

Chiral-at-metal catalysts: history, terminology, design, synthesis, and applications.

For decades, advances in chiral transition metal catalysis have been closely tied to the development of customized chiral ligands. Recently, however, an alternative approach to this traditional metal-plus-chiral-ligand method has emerged. In this new strategy, chiral transition metal catalysts are composed entirely of achiral ligands, with the overall chirality originating exclusively from a stereogenic metal center. This "chiral-at-metal" approach offers the benefit of structural simplicity. More importantly, by removing the need for chiral elements within the ligand framework, it opens up new possibilities for designing innovative catalyst architectures with unique properties. As a result, chiral-at-metal catalysis is becoming an increasingly important area of research. This review offers a comprehensive overview and detailed insights into asymmetric chiral-at-metal catalysis, encouraging scientists to explore new avenues in asymmetric transition metal catalysis and driving innovation in both fundamental and applied research.

Ds-FCRN: three-dimensional dual-stream fully convolutional residual networks and transformer-based global-local feature learning for brain age prediction.

The brain undergoes atrophy and cognitive decline with advancing age. The utilization of brain age prediction represents a pioneering methodology in the examination of brain aging. This study aims to develop a deep learning model with high predictive accuracy and interpretability for brain age prediction tasks. The gray matter (GM) density maps obtained from T1 MRI data of 16,377 healthy participants aged 45 to 82 years from the UKB database were included in this study (mean age, , 7811 men). We propose an innovative deep learning architecture for predicting brain age based on GM density maps. The architecture combines a 3D dual-stream fully convolutional residual network (ds-FCRN) with a Transformer-based global-local feature learning paradigm to enhance prediction accuracy. Moreover, we employed Shapley values to elucidate the influence of various brain regions on prediction precision. On a test set of 3,276 healthy subjects (mean age, , 1561 men), our 3D ds-FCRN model achieved a mean absolute error of 2.2 years in brain age prediction, outperforming existing models on the same dataset. The posterior interpretation revealed that the temporal lobe plays the most significant role in the brain age prediction process, while frontal lobe aging is associated with the greatest number of lifestyle factors. Our designed 3D ds-FCRN model achieved high predictive accuracy and high decision transparency. The brain age vectors constructed using Shapley values provided brain region-level insights into life factors associated with abnormal brain aging.

RCFormer: radiation correction method for degraded multispectral UAV images using vision transformers

ABSTRACT In modern remote sensing technology, Unmanned Aerial Vehicles (UAVs) have become crucial data acquisition platforms. However, low-altitude remote sensing images from UAVs are often affected by atmospheric scattering and absorption, particularly under conditions like haze, which severely degrade image quality. Traditional radiation correction methods cannot accurately capture the true reflectance of ground objects under these conditions. This paper introduces RCFormer, an innovative Transformer-based network architecture. It combines the self-attention mechanism of Swin Transformer with a novel network structural design, especially incorporating parallel convolutional structures to enhance the extraction capability of surface features. Additionally, with the integration of atmospheric characteristic refinement module (ACRM) and Multi-scale Feature Fusion Gate (MFFG), the proposed RCFormer not only captures atmospheric information from deep features but also effectively fuses features across different scales, thereby improving the accuracy, transferability and robustness of the radiation correction model. Finally, quantitative and qualitative evaluations using real-world data demonstrate that RCFormer outperforms other dehazing networks in both degraded and haze-free images while also offering good efficiency and ease of operation.

Advantages and Challenges of Graphene Transistors for High-Frequency Applications

Graphene is regarded as an ideal material for next-generation high-frequency (HF) electronic devices, attributed to its unique two-dimensional structure, high carrier mobility, and exceptional electrical properties. In recent years, the limitations of conventional silicon-based technologies for high-frequency applications have become increasingly apparent, resulting in heightened interest in graphene transistors, which exhibit significant potential for applications in wireless communications, radar systems, and terahertz technology. These transistors exhibit high cutoff frequency and low noise characteristics, along with favorable bipolar properties, which provide them with significant advantages in radio frequency (RF) applications. However, despite the significant theoretical advantages of graphene materials, their practical application is still restricted by multiple factors. Specifically, insufficient voltage gain from weak current saturation, material uniformity issues, and large-scale production challenges have restricted the widespread adoption of graphene transistors in high-frequency electronic devices. This paper reviews the relevant literature, examines the disparity between the electrical performance and practical applications of graphene transistors, and identifies the key factors influencing their implementation. The results demonstrate that targeted bandgap engineering, innovative device architectures, and the exploration of new materials can help graphene transistors can mitigate existing challenges and position themselves as key components in future HF electronic devices, driving significant advancements and broader integration of related technologies.

Iron‐Nitrogen‐Carbon Aerogel for Enhanced Oxygen Reduction in Acidic Media: The Influence of Temperature

AbstractThe oxygen reduction reaction (ORR) in an acidic environment is crucial for fuel cell technology. Understanding its complex kinetics and developing advanced catalyst materials have the potential to drive significant improvements in energy efficiency, paving the way for sustainable, green energy solutions. In this work, Iron‐Nitrogen‐Carbon@aerogel (Fe(FcP)x‐N‐C@Aerogel) catalysts are developed by carbonizing polypyrrole (PPy) and ferrocene. The ORR performance of these catalysts is investigated across different annealing temperatures. The catalysts’ shape and structure are validated through scanning electron microscopy (SEM) and X‐ray photoelectron spectroscopy (XPS), revealing that changes in annealing temperature affect morphology and nitrogen‐containing functional groups of the catalyst. Linear sweep voltammetry (LSV) and rotating disk electrode (RDE) studies demonstrated that Fe(FcP)800‐N‐C@Aerogel catalysts exhibit excellent performance, with a half‐wave potential of 0.687 V and an average electron transfer number of 3.98 under acidic conditions. These findings suggest a near‐four‐electron reaction pathway, highlighting the catalyst's strong ORR activity, high efficiency, and durability, with only 17.4 mV LSV curve decay after 10,000 cycles. In conclusion, Fe(FcP)x‐N‐C@Aerogel advances ORR catalysis in acidic media by delivering exceptional performance and durability, driven by its innovative architecture and precisely engineered active sites, setting a new benchmark for high‐efficiency energy conversion.

A Novel 5G Fronthaul Architecture Based on Quantum Security Protection

In the digital age, 5G technology has significantly enhanced global communication capabilities, providing strong support for various industries. However, 5G falls short of achieving comprehensive intelligence within the Internet of Things (IoT), propelling the development of 6G technology. 6G is expected to further enhance network performance by integrating advanced technologies such as AI and quantum communication, while expanding application scenarios to include communication in extreme environments for comprehensive global connectivity. This paper proposes an innovative fronthaul architecture that applies Quantum Key Distribution (QKD) technology to the 5G fronthaul, leveraging its unconditionally secure key generation and distribution mechanism. In this design, costlier Alice devices are placed on the AAU side, while less expensive Bob devices are positioned on the DU side, optimizing deployment to reduce engineering costs and lower deployment barriers. The feasibility of this architecture is demonstrated by calculating the secure key rate. Comparative analysis with existing research is performed to clarify future research directions. These innovations not only enhance 5G network security but also offer new solutions for the security requirements of 6G networks, such as data security, network resilience, and algorithm transparency. Our research offers strategic value for future network security, laying the groundwork for reliable networks.

Leveraging deep learning for robust EEG analysis in mental health monitoring.

Mental health monitoring utilizing EEG analysis has garnered notable interest due to the non-invasive characteristics and rich temporal information encoded in EEG signals, which are indicative of cognitive and emotional conditions. Conventional methods for EEG-based mental health evaluation often depend on manually crafted features or basic machine learning approaches, like support vector classifiers or superficial neural networks. Despite the potential of these approaches, they often fall short in capturing the intricate spatiotemporal relationships within EEG data, leading to lower classification accuracy and poor adaptability across various populations and mental health scenarios. To overcome these limitations, we introduce the EEG Mind-Transformer, an innovative deep learning architecture composed of a Dynamic Temporal Graph Attention Mechanism (DT-GAM), a Hierarchical Graph Representation and Analysis (HGRA) module, and a Spatial-Temporal Fusion Module (STFM). The DT-GAM is designed to dynamically extract temporal dependencies within EEG data, while the HGRA models the brain's hierarchical structure to capture both localized and global interactions among different brain regions. The STFM synthesizes spatial and temporal elements, generating a comprehensive representation of EEG signals. Our empirical results confirm that the EEG Mind-Transformer significantly surpasses conventional approaches, achieving an accuracy of 92.5%, a recall of 91.3%, an F1-score of 90.8%, and an AUC of 94.2% across several datasets. These findings underline the model's robustness and its generalizability to diverse mental health conditions. Moreover, the EEG Mind-Transformer not only pushes the boundaries of state-of-the-art EEG-based mental health monitoring but also offers meaningful insights into the underlying brain functions associated with mental disorders, solidifying its value for both research and clinical settings.

Fractal-domain Transformer Based on Learnable Multifractal Spectrum for Chaotic Systems Classification

Conventional deep learning in the spatiotemporal-frequency domain frequently encounter challenges in terms of slow convergence rates and limited generalization, particularly for classification of chaotic systems. To address these limitations, this paper introduces a novel fractal-inspired deep network model, specifically, the Multifractal Spectrum Transformer (MFS-Transformer), grounded in learnable multifractal analysis. Initially, we put forward the conceptual framework of fractal learning, compared with traditional fractal signal processing methodologies and spatial-temporal domain learning paradigms. Subsequently, a Learnable Multifractal Spectrum (LMFS) derived from 3D spatial gridding, coupled with fractal-domain filtering, is proposed to construct the iterative learning process within the fractal domain. Further, we formulate the MFS-Transformer, an innovative architecture that integrates multi-channel embedding, LMFS, fractal-domain filtering, residual fusion mechanisms, a mixer module, and a classifier, tailored for chaotic system classification. Ultimately, we evaluate the efficacy of our model in classifying 3D chaotic systems under stringent conditions of short-term sequences and low Signal-to-Noise Ratio (SNR). Experimental outcomes underscore the substantial performance gains achieved by the MFS-Transformer, with classification accuracy enhancements of 13.34% and 5.00% over existing Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), respectively, under SNR=0dB and 32-sample sequences. These findings validate the superiority of the MFS-Transformer in addressing the complexities of chaotic system classification under complex scenarios. This research not only advances the frontier of fractal deep learning but also presents a novel perspective and methodology for tackling intricate spatiotemporal classification problems.

Open-Source Framework for Sizing Hybrid and Electric General Aviation Aircraft

This paper presents FAST-OAD-CS23-HE, an open-source framework that has been developed as an extension of the existing FAST-OAD-GA framework to allow for medium fidelity sizing of hybrid and electric aircraft using component models dependent on operating conditions and sizing criteria. It inherits Overall Aircraft Design methodologies from the original framework and adds a library of models to represent physical components for hybrid powertrains. It also adds a generic methodology that enables the extension to multiphysic simulation for the powertrain and the consideration of synergistic interactions and supports the addition of new components. A graph-based description of the powertrain was chosen to easily describe complex powertrains that could be considered in innovative architectures as well as ease the future interfacing with external tools. In addition to that, the graph-based approach allows the automation of the construction of the design problem, which removes the need for users to handle complex scripts. It has been developed after a comparison of existing methodologies and open-source framework in an effort to bridge the gap in terms of preliminary design of electric aircraft. With the models implemented with the default delivery of the code, two aircraft were studied to serve as a reference to showcase the capabilities of this framework.

Automated Chest X-Ray Diagnosis Report Generation with Cross-Attention Mechanism

In the medical field, it is extremely important to use deep learning technology to automatically generate diagnostic reports for chest X-ray images. This technology provides an effective solution to the challenges faced by the medical field in processing large numbers of chest X-ray images. Especially during large-scale outbreaks of epidemics such as the new COVID-19, rapid and accurate screening and diagnosis of cases become important tasks. This study uses deep learning technology to automatically generate diagnostic reports for chest X-ray images, which significantly reduces the workload of doctors, reduces the risk of misdiagnosis and missed diagnosis, and provides technical support for improving public health emergency response capabilities. In this study, we propose an innovative network architecture to address the limitations of traditional image description networks in generating chest X-ray diagnostic reports, especially the large area deviation between abnormal and normal areas, and the lack of effective alignment of the two modalities of image and text. The convolutional block attention module (CBAM) is adopted to effectively alleviate the data bias problem through a sophisticated feature attention mechanism and improve the model’s ability to recognize abnormal image areas. The cross-attention mechanism is adopted to optimize the alignment process between images and texts, ensuring the accuracy and reliability of the diagnosis report.

Innovative mechanical sensors architectures and their physics applications

Innovative mechanical sensors architectures and their physics applications

Emerging 2D materials hardware for in-sensor computing.

The advent of the novel in-sensor/near-sensor computing paradigm significantly eliminates the need for frequent data transfer between sensory terminals and processing units by integrating sensing and computing functions into a single device. This approach surpasses the traditional configuration of separate sensing and processing units, thereby greatly simplifying system complexity. Two-dimensional materials (2DMs) show immense promise for implementing in-sensor computing systems owing to their exceptional material properties and the flexibility they offer in designing innovative device architectures with heterostructures. This review highlights recent progress and advancements in 2DM-based in-sensor computing research, summarizing the unique physical mechanisms that can be leveraged in 2DM-based devices to achieve sensory responses and the essential biomimetic synaptic characteristics for computing functions. Additionally, the potential applications of 2DM-based in-sensor computing systems are discussed and categorized. This review concludes with a perspective on future development directions for 2DM-based in-sensor computing.

Advanced MASnI3 and PTAA-Integrated ZnO2 Perovskite Composite: Optimizing Stability and Charge Dynamics for Next-Gen Photobatteries.

To advance off-grid energy solutions, developing flexible photobatteries capable of direct light charging is essential. This study presents an innovative photobattery architecture that incorporates zinc oxide (ZnO2) as an electron-transporting and hole-blocking layer, combined with a hybrid methylammonium tin iodide composite with poly-triarylamine (MASnI3/PTAA) for light absorption and hole transport. PTAA facilitates efficient hole transport to the anode, thereby enhancing charge separation and reducing recombination losses. The MASnI3 perovskite serves as an effective sunlight absorber, generating charge carriers. ZnO2, known for its high chemical stability and rapid electron mobility, effectively blocks holes and ensures swift electron flow to the cathode, which optimizes the overall charge transfer dynamics. This refined structure achieves a photoconversion efficiency enhancement of up to 0.53% and retains approximately 98% of its capacity after 700 cycles. The optimized MASnI3/PTAA/ZnO2 photobattery demonstrates a 3-fold reduction in light charging time, positioning it as a strong candidate for practical, light-rechargeable energy storage applications.

Innovative modified-net architecture: enhanced segmentation of deep vein thrombosis

A new era for diagnosing and treating Deep Vein Thrombosis (DVT) relies on precise segmentation from medical images. Our research introduces a novel algorithm, the Modified-Net architecture, which integrates a broad spectrum of architectural components tailored to detect the intricate patterns and variances in DVT imaging data. Our work integrates advanced components such as dilated convolutions for larger receptive fields, spatial pyramid pooling for context, residual and inception blocks for multiscale feature extraction, and attention mechanisms for highlighting key features. Our framework enhances precision of DVT region identification, attaining an accuracy of 98.92%, with a loss of 0.0269. The model also validates sensitivity 96.55%, specificity 96.70%, precision 98.61%, dice 97.48% and Intersection over Union (IoU) 95.10% offering valuable insights into DVT segmentation. Our framework significantly improves segmentation performance over traditional methods such as Convolutional Neural Network , Sequential, U-Net, Schematic. The management of DVT can be improved through enhanced segmentation techniques, which can improve clinical observation, treatment planning, and ultimately patient outcomes.

ARCHITECTURAL INNOVATIONS OF THE MEDIEVAL ERA THROUGH STRUCTURAL AND MATERIAL ADVANCEMENTS

The medieval period was a time of great architectural innovation, largely due to structural design and material technology advances. From roughly the 5th through the 15th century, new ways in constructing buildings greatly impacted the current face of Europe and the Middle East. Some of these innovations included the invention and use of pointed arches, ribbed vaults, and flying buttresses, which allowed for the construction of soaring cathedrals and fortified castles. These features permitted even more excellent stability and height of buildings, making designs far more complex and involved with the character-defining features of Gothic architecture. Besides this, various building materials came into play, such as stone, mortar, and early forms of reinforced wood that enabled architects and builders to experiment with larger spans and complicated façades. While the architectural novelties of the time reflected social and cultural influences, churches, palaces, and civic buildings alike came to be manifestations of power, faith, and communal identity. The subsequent article discusses how the structural innovations coupled with choices of material contributed toward the singular architectural achievements of the medieval period and how the technical ingenuity laid the cornerstone for future architectural practices.

Migrating from Developing Asynchronous Multi-Threading Programs to Reactive Programs in Java

Modern software application development imposes standards regarding high performance, scalability, and minimal system latency. Multi-threading asynchronous programming is one of the standard solutions proposed by the industry for achieving such objectives. However, the recent introduction of the reactive programming interface in Java presents a potential alternative approach for addressing such challenges, promising performance improvements while minimizing resource utilization. The research examines the migration process from the asynchronous paradigm to the reactive paradigm, highlighting the implications, benefits, and challenges resulting from this transition. To this end, the architecture, technologies, and design of a support application are presented, outlining the practical aspects of this experimental process while closely monitoring the phased migration. The results are examined in terms of functional equivalence, testing, and comparative analysis of response times, resource utilization, and throughput, as well as the cases where the reactive paradigm proves to be a solution worth considering. Across multiple scenarios, the reactive paradigm demonstrated advantages such as up to 12% reduction in memory usage, 56% faster 90th percentile response times, and a 33% increase in throughput under high-concurrency conditions. However, the results also reveal cases, such as data-intensive scenarios, where asynchronous programming outperforms reactive approaches. Additionally, possible directions for further research and development are presented. This paper not only investigates the design and implementation process but also sets a foundation for future research and innovation in dependable systems, collaborative technologies, sustainable solutions, and distributed system architecture.

A 0.73 dB Multi-Gain Low Noise Amplifier Design with Fast Mode-Switching for 5G/4G Applications.

In this paper, a sub-1dB Low Noise Amplifier (LNA) with several gain modes, including amplification and attenuation modes required for the fifth and fourth generations (5G/4G) of mobile network applications, is proposed. Its current consumption is adaptive for every gain mode and varies to lower currents for lower amplifications due to the importance of current consumption for mobile network applications. The proposed LNA features an innovative architecture with a three-core input structure supporting multi-gain modes, achieving high gain and ultra-low noise performance. Additionally, the design integrates a cascade switching mechanism to ensure fast transitions between the gain modes and maintain operational stability. A reconfigurable input structure is introduced to support multiple input stages, enabling the proposed LNA to be compatible with both 5G and 4G applications. The proposed design demonstrates the implementation of seven distinct gain modes with a maximum current consumption of 11.68 mA, achieving proper input matching in each gain mode. The LNA delivers a maximum gain of 20.4 dB with a noise figure of 0.73 dB. Moreover, the most stringent mode switching condition achieved, the ON time, is as short as 1.295 µs, and the gain mode transition speed is an impressive 0.874 µs, ensuring extremely fast mode transitions. The proposed LNA occupies an area of 700 µm × 500 µm and is fabricated using a 65 nm FD-SOI process.

Cost vs. Performance: Raspberry Pi and ESP32 in Face Detection Tasks

A cost-effective, scalable solution that leverages low-cost embedded systems for computationally intensive tasks like face detection is critical for the growing demands of IoT applications, motivating the development of innovative hybrid architectures. This paper proposes such a framework, combining the ESP32 CAM as a distributed image capture unit with light preprocessing and the Raspberry Pi as a centralized processing unit. By addressing the limitations of standalone implementations—low computational capability of the ESP32 CAM and high deployment costs of the Raspberry Pi—our framework achieves a balance between affordability and performance. We present the analysis of the devices, evaluating metrics such as frame rate, detection accuracy, and cost-efficiency. The results highlight the potential of the hybrid system to significantly lower costs while enabling scalable, real-time face detection in IoT scenarios. This study contributes to the ongoing research by proposing an adaptable, resource-optimized framework suitable for diverse use cases, from smart surveillance to retail monitoring, paving the way for more efficient IoT-based vision systems.

Species mass transfer governs the selectivity of gas diffusion electrodes toward H2O2 electrosynthesis

The meticulous design of advanced electrocatalysts and their integration into gas diffusion electrode (GDE) architectures is emerging as a prominent research paradigm in the H2O2 electrosynthesis community. However, it remains perplexing that electrocatalysts and assembled GDE frequently exhibit substantial discrepancies in H2O2 selectivity during bulk electrolysis. Here, we elucidate the pivotal role of mass transfer behavior of key species (including reactants and products) beyond the intrinsic properties of the electrocatalyst in dictating electrode-scale H2O2 selectivity. This tendency becomes more pronounced in high reaction rate (current density) regimes where transport limitations are intensified. By utilizing diffusion-related parameters (DRP) of GDEs (i.e., wettability and catalyst layer thickness) as probe factors, we employ both short- and long-term electrolysis in conjunction with in-situ electrochemical reflection-absorption imaging and theoretical calculations to thoroughly investigate the impact of DRP and DRP-controlled local microenvironments on O2 and H2O2 mass transfer. The mechanistic origins of diffusion-dependent conversion selectivity at the electrode scale are unveiled accordingly. The fundamental insights gained from this study underscore the necessity of architectural innovations for mainstream hydrophobic GDEs that can synchronously optimize mass transfer of reactants and products, paving the way for next-generation GDEs in gas-consuming electroreduction scenarios.