Evaluation Of Architecture Research Articles

Exploring biometric domain adaptation in human action recognition models for unconstrained environments

AbstractIn conventional machine learning (ML), a fundamental assumption is that the training and test sets share identical feature distributions, a reasonable premise drawn from the same dataset. However, real-world scenarios often defy this assumption, as data may originate from diverse sources, causing disparities between training and test data distributions. This leads to a domain shift, where variations emerge between the source and target domains. This study delves into human action recognition (HAR) models within an unconstrained, real-world setting, scrutinizing the impact of input data variations related to contextual information and video encoding. The objective is to highlight the intricacies of model performance and interpretability in this context. Additionally, the study explores the domain adaptability of HAR models, specifically focusing on their potential for re-identifying individuals within uncontrolled environments. The experiments involve seven pre-trained backbone models and introduce a novel analytical approach by linking domain-related (HAR) and domain-unrelated (re-identification (re-ID)) tasks. Two key analyses addressing contextual information and encoding strategies reveal that maintaining the same encoding approach during training results in high task correlation while incorporating richer contextual information enhances performance. A notable outcome of this study is the comprehensive evaluation of a novel transformer-based architecture driven by a HAR backbone, which achieves a robust re-ID performance superior to state-of-the-art (SOTA). However, it faces challenges when other encoding schemes are applied, highlighting the role of the HAR classifier in performance variations.

Sixth Generation 6G to the Waying Seventh 7G Wireless Communication Visions and Standards, Challenges, Applications

The increasing need for next-generation wireless networks becomes apparent as the Internet of Everything (IoE) gains prominence in smart services, projecting widespread popularity in the future. While sixth generation (6G) networks have the capability to support a diverse range of IoE services, their potential limitations in meeting the demands of innovative applications prompt consideration for seventh generation (7G) wireless systems. This article seeks to compare the characteristics of 6G and the planned 7G wireless systems. The commercial development of fifth generation (5G) mobile communication systems is currently underway, introducing new services and enhancing user experiences. Despite these advancements, 5G encounters challenges that require on-going improvements. With the International Telecommunication Union Radio communication Sector (ITU-R) actively envisioning 6G and anticipating a consensus on 7G by October 2024, numerous unresolved questions persist in global discussions. This paper delivers a comprehensive overview of the current understanding of 7G, exploring the vision, technical requirements, and application possibilities. It presents a critical evaluation of the 7G network architecture and essential technologies. Furthermore, the article provides an in-depth examination of advanced 7G verification platforms and existing test beds, unveiling these aspects for the first time. Future research directions and lingering issues are emphasized to contribute to the on-going global discourse on 7G networks. Discussions on lessons learned from 7G networks culminate in the suggestion that 7G systems signify the pinnacle of mobile communication technology. The ongoing research on 7G, an intelligent cellular technology, holds promise for the future of wireless communication.

Evaluation of sleep habits and sleep architecture in children referred to the sleep ward of Qazvin children's hospital, Qazvin, Iran during 2016-2022

Evaluation of sleep habits and sleep architecture in children referred to the sleep ward of Qazvin children's hospital, Qazvin, Iran during 2016-2022

Abstract PR016: Topology of the tumor microenvironment: Integrating imaging, modeling, and topological data analysis

Abstract Cancer evolution involves multiple stochastic processes across many scales. In addition to genomic and transcriptomic changes, hallmarks of cancer within solid tumors include the malignant co-evolution of the tumor microenvironment. Although the evaluation of tissue architecture plays a key role in pathologist staging, the automated extraction of additional structural features remains far from translation to the clinic. However, one particularly effective novel methodology for extracting patterns from these clinical images is topological data analysis(TDA). TDA is a field that studies the shape of data and its connectedness using concepts from geometric and algebraic topology. Recent work using topological data analysis has allowed the extraction of image features. Such features have been associated with clinical outcomes in both the analysis of H&E images in prostate cancer and CT images in lung cancer. We set out to question how the topological features of tissues, including tumor and stromal composition, evolve during treatment and between original sites and metastasis. To analyze this we use a combination of tissue sections, imaging, and cell-automata models and persistent homology and cubical persistence techniques from TDA. We apply certain topological methods for the first time to tissue images (H&E) and to the output of spatial simulations. We focus specifically on analyzing the evolution of topological features in ovarian and lung cancer and compare our features to 2D and 3D simulations (HAL) with varying game theoretical interactions as well as comparing features across tissues from matched primary and metastatic sites. We find that several topological features are conserved between primary and metastatic sites and some feature evolution is replicated in simulation. Citation Format: Rowan Barker-Clarke, Jason M. Gray, Jacob G. Scott. Topology of the tumor microenvironment: Integrating imaging, modeling, and topological data analysis [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Cancer Evolution and Data Science: The Next Frontier; 2023 Dec 3-6; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_2):Abstract nr PR016.

Quality measures for the evaluation of machine learning architectures on the quantification of epistemic and aleatoric uncertainties in complex dynamical systems

Machine learning methods for the construction of data-driven reduced order model models are used in an increasing variety of engineering domains, especially as a supplement to expensive computational fluid dynamics for design problems. An important check on the reliability of surrogate models is Uncertainty Quantification (UQ), a self assessed estimate of the model error. Accurate UQ allows for cost savings by reducing both the required size of training data sets and the required safety factors, while poor UQ prevents users from confidently relying on model predictions. We introduce two quality measures for the quantification of uncertainty that are suitable for high dimensional problems: the distribution of normalized residuals on validation data and the distribution of estimated uncertainties. While the distribution of estimated uncertainties evaluates how confident the model is in making predictions, the distribution of normalized residuals evaluates how accurate these uncertainty estimates are We examine several machine learning techniques, including both Gaussian processes and a family UQ-augmented neural networks: Ensemble neural networks (ENN), Bayesian neural networks (BNN), Dropout neural networks (D-NN), and Gaussian neural networks (G-NN). We evaluate UQ accuracy (distinct from model accuracy) We apply these metrics to two model data sets representative of complex dynamical systems: an ocean engineering problem in which a ship traverses irregular wave episodes, and a dispersive wave turbulence system with extreme events, the Majda–McLaughlin–Tabak model. We present conclusions concerning model architecture and hyperparameter tuning for use in applications of UQ to dynamical systems, such as active learning schemes.

Vessel trajectory prediction with recurrent neural networks: An evaluation of datasets, features, and architectures

Maritime situational awareness tasks such as port management, collision avoidance, and search-and-rescue missions rely on accurate knowledge of vessel locations. The availability of historical vessel trajectory data through the Automatic Identification System (AIS) has enabled the development of prediction methods, with a recent focus on trajectory prediction via recurrent neural networks (RNNs) and other deep learning architectures. While these methods have shown promising performance benefits over kinematic and clustering-based models, comparing among RNN-based models remains difficult due to variations in evaluation datasets, region sizes, vessel types, and numerous other design choices. As a result, it is not clear whether recent methods based on highly-sophisticated network architectures are necessary to achieve strong prediction performance. In this work, we present a simple fusion-based RNN approach to vessel trajectory prediction that allows for easy incorporation of exogenous variables. We perform an extensive ablation study to measure the impact of various modeling choices, including preprocessing, loss functions, and the choice of features, as well as the first usage of surface current information in vessel trajectory prediction. We demonstrate that our approach achieves state-of-the-art performance on three large regions off the United States coast, obtaining an improvement of up to 0.88 km over competing methods when predicting three hours into the future. We conclude that our simple architecture can outperform more complicated architectures while incurring a lower memory cost. Further, we show that the choice of loss function and the inclusion of surface current information both have significant impact on prediction performance.

Assessment of 5G Connectivity for UAV Operations in Urban Environments: An Analysis of Network Performance and UAV Communication Architecture

This article presents an in-depth analysis of the integration of 5G connectivity for unmanned aerial vehicle (UAV) operations in urban environments, focusing on network performance and the architectural framework necessary for optimal UAV communication. As UAVs become increasingly prevalent in urban applications, from surveillance to delivery services, the demand for robust and reliable communication systems has become paramount. This study explores the unique challenges posed by urban settings, such as signal obstruction and high-density network traffic, and examines how 5G technology offers solutions through advanced features like beamforming. By detailing the setup and evaluation of a UAV communication architecture that incorporates 5G modules, including the use of the NVIDIA Jetson AGX Orin Developer Kit for real-world testing, this article highlights the significant improvements in data rate, latency, and reliability achieved with 5G networks compared to traditional communication technologies. The findings underscore the transformative potential of 5G connectivity in enhancing UAV operational efficiency and reliability in complex urban landscapes. Through case studies and practical insights, the article also provides a glimpse into the future of urban UAV applications, paving the way for further research and development in this dynamic field.

Data Management across Geographically-Distributed Autonomous Systems: Architecture, Implementation, and Performance Evaluation

Supporting efficient geographically-distributed autonomous data management is one of the critical obstacles for opening up a big data era. It creates the need for investigating such kind of systems in big data environments. In this paper, a distributed autonomous data management system is put forward, exhibiting the following features. (1) A distributed architecture designed to meet the requirements of autonomous data management by allowing interconnection, intercommunication, and interoperation of multiple sites over the Internet. (2) An autonomous, multi-level, unstructured data storage system to meet high-efficiency storage needs, with reference to the distributed heterogeneous data storage theories. (3) A distributed autonomous data indexing and retrieval system to support metadata search & full-text searching, fast loading, remote access, and unified view. Experimental and industrial application results demonstrate that the proposed system has high potential to reduce access time and improve storage efficiency, while maintaining satisfactory availability and scalability.

Similarity surrogate-assisted evolutionary neural architecture search with dual encoding strategy

<abstract><p>Neural architecture search (NAS), a promising method for automated neural architecture design, is often hampered by its overwhelming computational burden, especially the architecture evaluation process in evolutionary neural architecture search (ENAS). Although there are surrogate models based on regression or ranking to assist or replace the neural architecture evaluation process in ENAS to reduce the computational cost, these surrogate models are still affected by poor architectures and are not able to accurately find good architectures in a search space. To solve the above problems, we propose a novel surrogate-assisted NAS approach, which we call the similarity surrogate-assisted ENAS with dual encoding strategy (SSENAS). We propose a surrogate model based on similarity measurement to select excellent neural architectures from a large number of candidate architectures in a search space. Furthermore, we propose a dual encoding strategy for architecture generation and surrogate evaluation in ENAS to improve the exploration of well-performing neural architectures in a search space and realize sufficiently informative representations of neural architectures, respectively. We have performed experiments on NAS benchmarks to verify the effectiveness of the proposed algorithm. The experimental results show that SSENAS can accurately find the best neural architecture in the NAS-Bench-201 search space after only 400 queries of the tabular benchmark. In the NAS-Bench-101 search space, it can also get results that are comparable to other algorithms. In addition, we conducted a large number of experiments and analyses on the proposed algorithm, showing that the surrogate model measured via similarity can gradually search for excellent neural architectures in a search space.</p></abstract>

Construction of network access Layer security protection System based on zero trust architecture

In order to further solve the problem of weak attack defense existing in the traditional network architecture, improve the shortcomings of network architecture in processing capacity, hardware expansion, security protection and other aspects, and better cope with the security challenges under the new network architecture, this article aims to provide more secure computer information exchange and sharing services for enterprises. A network access layer security protection system based on zero trust architecture is constructed. At the same time, the zero-trust architecture evaluation algorithm based on fuzzy theory is used to construct the security protection model. System tests show that more than 100 hosts on the enterprise access layer network use switches that support SNMP, and the security protection system at the network access layer processes more than 1500 attacks. In addition, malicious IP address changes are effectively prevented and the security protection level is greatly improved.

SCAG: A Stratified, Clustered, and Growing-Based Algorithm for Soybean Branch Angle Extraction and Ideal Plant Architecture Evaluation.

Three-dimensional (3D) phenotyping is important for studying plant structure and function. Light detection and ranging (LiDAR) has gained prominence in 3D plant phenotyping due to its ability to collect 3D point clouds. However, organ-level branch detection remains challenging due to small targets, sparse points, and low signal-to-noise ratios. In addition, extracting biologically relevant angle traits is difficult. In this study, we developed a stratified, clustered, and growing-based algorithm (SCAG) for soybean branch detection and branch angle calculation from LiDAR data, which is heuristic, open-source, and expandable. SCAG achieved high branch detection accuracy (F-score = 0.77) and branch angle calculation accuracy (r = 0.84) when evaluated on 152 diverse soybean varieties. Meanwhile, the SCAG outperformed 2 other classic algorithms, the support vector machine (F-score = 0.53) and density-based methods (F-score = 0.55). Moreover, after applying the SCAG to 405 soybean varieties over 2 consecutive years, we quantified various 3D traits, including canopy width, height, stem length, and average angle. After data filtering, we identified novel heritable and repeatable traits for evaluating soybean density tolerance potential, such as the ratio of average angle to height and the ratio of average angle to stem length, which showed greater potential than the well-known ratio of canopy width to height trait. Our work demonstrates remarkable advances in 3D phenotyping and plant architecture screening. The algorithm can be applied to other crops, such as maize and tomato. Our dataset, scripts, and software are public, which can further benefit the plant science community by enhancing plant architecture characterization and ideal variety selection.

System Architecture of Electronic Asset Supply Chain Intelligent Platform for Digital Higher Education

This research aims to design an intelligent platform architecture for electronic asset supply chains for digital higher education and to evaluate the architecture of the intelligent platform for electronic asset supply chains for digital higher education. The sample group consists of evaluations of the intelligent platform architecture for the electronic asset supply chains for digital higher education by experts. These experts assess and certify the appropriateness of the architecture, evaluating the content's suitability and the management processes. The evaluations were conducted by 5 experts who have experience in managing assets in higher education or relevant areas. The research results indicate that the designed intelligent platform architecture for electronic asset supply chains for digital higher education, on average, scored 4.43, which is considered 'good'. The evaluation of its developmental trend from architecture to platform has an average score of 4.80, considered 'very good'. Following that, both the system (Administrators) and the (Webserver and Database Server) evaluations yielded the same average score of 4.60, which is also ranked as 'very good.'

Comparing age- and bone-related differences in collagen fiber orientation: A case study of bats and laboratory mice using quantitative polarized light microscopy.

As bones age in most mammals, they typically become more fragile. This state of bone fragility is often associated with more homogenous collagen fiber orientations (CFO). Unlike most mammals, bats maintain mechanically competent bone throughout their lifespans, but little is known of positional and age-related changes in CFO within wing bones. This study tests the hypothesis that age-related changes in CFO in big brown bats (Eptesicus fuscus) differ from those of the standard mammalian model for skeletal aging, the C57BL/6 laboratory mouse. We used data from quantitative polarized light microscopy (qPLM) to compare CFO across the lifespan of long-lived big brown bats and age matched C57BL/6 mice. Eptesicus and C57BL/6 mice displayed idiosyncratic patterns of CFO. Consistent age-related changes were only apparent in the outer cortical bone of Eptesicus, where bone tissue is more longitudinally arranged and more anisotropic in older individuals. Both taxa displayed a ring of more transversely oriented bone tissue surrounding the medullary cavity. In Eptesicus, this tissue represents a greater proportion of the overall cross-section, and is more clearly helically aligned (arranged at 45° to the bone long axis) than similar bone tissue in mice. Bat wing bones displayed a proximodistal gradient in CFO anisotropy and longitudinal orientation in both outer and inner cortical bone compartments. This study lays a methodological foundation for the quantitative evaluation of bone tissue architecture in volant and non-volant mammals that may be expanded in the future.

Evaluation of SSD Architecture for Small Size Object Detection: A Case Study on UAV Oil Pipeline Monitoring

Oil pipeline monitoring using Unmanned Airborne Vehicles (UAV) can be done by utilizing Deep Learning. Deep Learning can be used to automatically detect harmed or unauthorized objects near the pipeline for further action by the authority. Input video in the pipeline area taken from the UAV has unique characteristics. It has low resolution with dense composition object in the image. The detected object also has a small scale as the objects are far away from the UAV. Thus, the selection of the Deep Learning algorithm is important to get a desirable result with the following conditions. Single Shot Multi-Box (SSD) is one of the popular Deep Learning algorithms with fast calculation compared to others and suitable for real-time object detection. Previous works on this topic using low to medium altitude dataset (20–200 m). This paper provides an evaluation of SSD implementation to detect vehicles on high-altitude dataset (300 m). As much as 2482 dataset is fed into SSD architecture and trained to detect 3 class of vehicles. The result shows the mAP and mAR are 0.026360 and 0.067377, respectively. However, the low lost function value shows that the model is able to classify the object correctly. In conclusion, the SSD cannot process low density information to correctly locate the object.

A hybrid Cycle GAN-based lightweight road perception pipeline for road dataset generation for Urban mobility.

One of the major problems that cause continual trouble in deep learning networks is that training a large network requires massive labelled datasets. The preparation of a massive labelled dataset is a cumbersome task and requires lot of human interventions. This paper proposes a novel generator network 'Sim2Real' transfer is a recent and fast-developing field in machine learning used to bridge the gap between simulated and real data. Training with simulated datasets often converges due to its size but fails to generalize real-world applications. Simulated datasets can be used to train and test deep learning models, enables the development and evaluation of new algorithms and architectures. By simulating road dataset, researchers can generate large amounts of realistic road-traffic dataset that can be used to study and understand several problems such as vehicular object tracking and classification, traffic situation analysis etc. The main advantage of such a transfer algorithm is to use the abundance of a simulated dataset to generate huge realistic-looking datasets to solve data-intense tasks. This work presents a novel, robust sim2real algorithm that converts the labels of a semantic segmentation map to a realistic-looking street view using the Cityscapes dataset and aims to achieve robust urban mobility for smart cities. Further, the generalizability of the Cycle Generative Adversarial Network (CycleGAN) architecture was tested by using an origami robot dataset for sim2real transfer. We show that the results were found to be qualitatively satisfactory for different traffic analysis applications. In addition, road perception was done using a lightweight SVM pipeline and evaluated on the KITTI dataset. We have incorporated Cycle Consistency Loss and Identity Loss as the metrics to evaluate the performance of the proposed Cycle GAN model. We inferred that the proposed Cycle GAN model provides an Identity loss of less than 0.2 in both the Cityscapes dataset and KITTI datasets. Also, we understand that the super-pixel resolution has a good impact on the quantitative results of the proposed Cycle GAN models.

Inflammatory pseudotumor of frontal sinus mimicking an aggressive malignancy: a rare case with diagnostic challenge on fine needle aspiration

Abstract Introduction/Objective Inflammatory pseudotumor (IPT) of sinonasal tract is a rare disease that sometimes mimicks malignancy in radiographic findings. Diagnosis of IPT on fine needle aspiration (FNA) is challenging due to limited cellularity and lack of architectural evaluation. Methods/Case Report Patient is a 59-year-old female presented with left front head swelling, facial pain and pressure. Imaging demonstrated a 3.0 cm hypermetabolic frontal sinus mass with osseous erosion through the calvaria table and deviating the adjacent dura, suspicious for malignancy. Also, an incidental 6.8 cm renal mass was identified and was proven to be clear cell renal cell carcinoma. FNA from frontal sinus showed loosely cohesive atypical cells with prominent nucleoli and occasional mitotic activity mixed with lymphoplasmacytic inflammation and eosinophils. Extensive immunostains showed that the lesional cells were only weakly stained for CD10 and desmin. Because of prominent cytologic atypia and radiological concerning for malignancy, FNA was diagnosed as malignancy not further classified. The patient underwent frontal sinus resection and the surgical specimen showed spindle cell proliferation surrounded by reactive woven bone with plump osteoblasts in a myxocollagenous stroma containing mixed lymphoplasmacytic cells and eosinophils. The spindle cells are positive for actin and calponin, negative for ALK. The spindle cells and plump osteoblasts corresponded to the atypical cells observed on FNA. The overall finding is consistent with IPT, although inflammatory myofibroblastic tumor cannot be completely ruled out. During follow up, the patient developed metastatic renal cell carcinoma in lung. Her frontal sinus lesion was stable for 17 months after surgery. Results (if a Case Study enter NA) NA Conclusion IPT of sinonasal tract is a rare lesion that sometimes demonstrate aggressive radiologic features. FNA from IPT poses a diagnostic challenge and can be misinterpreted as malignancy due to prominent cytologic atypia. Being aware of this rare entity helps avoid misdiagnosis and triggers the request for core biopsy during onsite evaluation.

FlowTransformer: A transformer framework for flow-based network intrusion detection systems

This paper presents the FlowTransformer framework, a novel approach for implementing transformer-based Network Intrusion Detection Systems (NIDSs). FlowTransformer leverages the strengths of transformer models in identifying the long-term behaviour and characteristics of networks, which are often overlooked by most existing NIDSs. By capturing these complex patterns in network traffic, FlowTransformer offers a flexible and efficient tool for researchers and practitioners in the cybersecurity community who are seeking to implement NIDSs using transformer-based models. FlowTransformer allows the direct substitution of various transformer components, including the input encoding, transformer, classification head, and the evaluation of these across any flow-based network dataset. To demonstrate the effectiveness and efficiency of the FlowTransformer framework, we utilise it to provide an extensive evaluation of various common transformer architectures, such as GPT 2.0 and BERT, on three commonly used public NIDS benchmark datasets. We provide results for accuracy, model size and speed. A key finding of our evaluation is that the choice of classification head has the most significant impact on the model performance. Surprisingly, Global Average Pooling, which is commonly used in text classification, performs very poorly in the context of NIDS. In addition, we show that model size can be reduced by over 50%, and inference and training times improved, with no loss of accuracy, by making specific choices of input encoding and classification head instead of other commonly used alternatives.

Architectural Evaluation of Healthcare Facilities: A Comprehensive Review and Implications for Building Design

Architectural Evaluation of Healthcare Facilities: A Comprehensive Review and Implications for Building Design

A new scale to assist in evaluating architectural proposals on the natural dimension based on psychometrics

Scientific and accurate evaluation of architecture is efficient for design optimization and resource conservation of sustainable cities. Complicated and multidisciplinary contents and architecture proposals are difficult to judge with validity and objectivity by current evaluation methods. A scale is a concentrated application of psychometrics that can collect expert knowledge objectively, scientifically, and comprehensively. This paper proposes the development of an architecture design evaluation scale on the natural dimension with the application of a web-based interface system. The evaluation items of the scale were determined by the Delphi process. Offering a broader dimension of information exchange, a web-based interface system was applied in the Delphi process and the scale pretest. Then, following psychometric research methodology, the pretest results were proven to be scientific and accurate by reliability and validity tests (KMO=0.72,P=1.06×10−105, Cronbach α=0.923). This research develops a new evaluation scale of architectural design, aiming to improve the quality of architectural design and the sustainability of cities.

An optimized visualization and quantitative protocol for in-depth evaluation of lymphatic vessel architecture in the liver.

The liver lymphatic system is essential for maintaining tissue fluid balance and immune function. The detailed structure of lymphatic vessels (LVs) in the liver remains to be fully demonstrated. The aim of this study is to reveal LV structures in normal and diseased livers by developing a tissue-clearing and coimmunolabeling protocol optimized for the tissue size and the processing time for three-dimensional (3-D) visualization and quantification of LVs in the liver. We showed that our optimized protocol enables in-depth exploration of lymphatic networks in the liver, consisting of LVs along the portal tract (deep lymphatic system) and within the collagenous Glisson's capsule (superficial lymphatic system) in different species. With this protocol, we have shown 3-D LVs configurations in relation to blood vessels and bile ducts in cholestatic mouse livers, in which LVs were highly dilated and predominantly found around highly proliferating bile ducts and peribiliary vascular plexuses in the portal tract. We also established a quantification method using a 3-D volume-rendering approach. We observed a 1.6-fold (P < 0.05) increase in the average diameter of LVs and a 2.4-fold increase (P < 0.05) in the average branch number of LVs in cholestatic/fibrotic livers compared with control livers. Furthermore, cholestatic/fibrotic livers showed a 4.3-fold increase (P < 0.05) in total volume of LVs compared with control livers. Our optimized protocol and quantification method demonstrate an efficient and simple liver tissue-clearing procedure that allows the comprehensive analysis of liver lymphatic system.NEW & NOTEWORTHY This article showed a comprehensive 3-D-structural analysis of liver lymphatic vessel (LV) in normal and diseased livers in relation to blood vessels and bile ducts. In addition to the LVs highly localized at the portal tract, we revealed capsular LVs in mouse, rat, and human livers. In cholestatic livers, LVs are significantly increased and dilated compared with normal livers. Our optimized protocol provides detailed spatial information for LVs remodeling in normal and pathological conditions.