IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-01-24 DOI: 10.1016/j.array.2026.100694

Bokai Li , Mingkang Guo , Yongli Jia , Tianzi Zeng , Xiaojing Liu

To address the issue of alert information overload in cloud platform monitoring, where unnecessary or duplicate alerts hinder the rapid identification of problem sources by operation and maintenance personnel, an automatic analysis system for cloud platform alert monitoring based on the random forest (RF) algorithm has been proposed. In the system architecture, the infrastructure layer creates multiple virtual machines through the CloudStack cloud platform, utilizing the C8051F0403 model chip as an information collector to acquire abnormal data. The core service layer, centered around the ARM7TDMI core microprocessor, designs the hardware structure of the monitoring terminal, integrating global GSM-based SMS transmission and reception to track abnormal operational states. The user interface layer supplies alert information to the system. The alert client is functionally designed by incorporating the random forest algorithm, which is capable of processing a large volume of alert log samples from the cloud platform system while avoiding overfitting. By constructing multiple decision trees, the algorithm enhances the accuracy of classification and regression tasks, effectively identifying and filtering out unnecessary or duplicate alert information, thereby enabling automated analysis of abnormal alert monitoring. Experimental results demonstrate that the system achieves effective noise reduction in alert data, maintains a low false alert rate in alert monitoring, and supports root-cause analysis of alerts. The application of this system can significantly mitigate alert overload, ensuring that the alert information received by operation and maintenance (O&M) personnel is more accurate and reliable, thereby facilitating quicker problem localization and effective resolution.

针对云平台监控中报警信息过载、不必要或重复报警阻碍运维人员快速识别问题来源的问题，提出了一种基于随机森林（RF）算法的云平台报警监控自动分析系统。在系统架构中，基础架构层通过CloudStack云平台创建多个虚拟机，利用C8051F0403型号芯片作为信息采集器采集异常数据。核心业务层以ARM7TDMI核心微处理器为核心，设计监控终端的硬件结构，集成基于全球gsm的短信收发，跟踪异常运行状态。用户界面层向系统提供警报信息。警报客户端在功能设计上结合随机森林算法，能够处理来自云平台系统的大量警报日志样本，同时避免过拟合。该算法通过构建多棵决策树，提高了分类和回归任务的准确性，有效地识别和过滤掉不必要或重复的警报信息，从而实现异常警报监测的自动化分析。实验结果表明，该系统对报警数据进行了有效的降噪，在报警监控中保持了较低的误报率，并支持对报警进行根本原因分析。该系统的应用可以显著缓解警报过载，保证运维人员接收到的警报信息更加准确可靠，从而更快地定位问题，有效地解决问题。

{"title":"Design of cloud platform alert monitoring and automatic analysis system based on random forest algorithm","authors":"Bokai Li , Mingkang Guo , Yongli Jia , Tianzi Zeng , Xiaojing Liu","doi":"10.1016/j.array.2026.100694","DOIUrl":"10.1016/j.array.2026.100694","url":null,"abstract":"<div><div>To address the issue of alert information overload in cloud platform monitoring, where unnecessary or duplicate alerts hinder the rapid identification of problem sources by operation and maintenance personnel, an automatic analysis system for cloud platform alert monitoring based on the random forest (RF) algorithm has been proposed. In the system architecture, the infrastructure layer creates multiple virtual machines through the CloudStack cloud platform, utilizing the C8051F0403 model chip as an information collector to acquire abnormal data. The core service layer, centered around the ARM7TDMI core microprocessor, designs the hardware structure of the monitoring terminal, integrating global GSM-based SMS transmission and reception to track abnormal operational states. The user interface layer supplies alert information to the system. The alert client is functionally designed by incorporating the random forest algorithm, which is capable of processing a large volume of alert log samples from the cloud platform system while avoiding overfitting. By constructing multiple decision trees, the algorithm enhances the accuracy of classification and regression tasks, effectively identifying and filtering out unnecessary or duplicate alert information, thereby enabling automated analysis of abnormal alert monitoring. Experimental results demonstrate that the system achieves effective noise reduction in alert data, maintains a low false alert rate in alert monitoring, and supports root-cause analysis of alerts. The application of this system can significantly mitigate alert overload, ensuring that the alert information received by operation and maintenance (O&M) personnel is more accurate and reliable, thereby facilitating quicker problem localization and effective resolution.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100694"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146073735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FORTRESS-FL: Byzantine-robust and privacy-preserving federated orchestration for next-generation networks 要塞- fl：拜占庭式健壮和保护隐私的下一代网络联合编排

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-02-02 DOI: 10.1016/j.array.2026.100680

Quang-Vinh Dang , Thi-Bich-Diem Vo , Ngoc-Son-An Nguyen

The transition to 6G and Open RAN (O-RAN) necessitates intelligent orchestration across multi-operator networks, yet this collaboration introduces severe security and privacy risks. Malicious operators may poison global models through adaptive attacks, while the exchange of raw gradients threatens data sovereignty. In this paper, we propose FORTRESS-FL, a robust and privacy-preserving federated learning framework designed for secure cross-domain orchestration. At its core is the TrustChain protocol, which synergizes a commit-then-reveal scheme to prevent adaptive manipulation, unsupervised spectral clustering for Byzantine detection, and a dynamic reputation system to isolate malicious actors. Furthermore, we integrate an adaptive Differential Privacy (DP) mechanism to rigorously protect operator data. Extensive evaluation on a real-world financial fraud dataset demonstrates that FORTRESS-FL achieves 100% detection accuracy against sign-flip attacks with 30% Byzantine adversaries, preventing the model divergence observed in standard baselines. Scalability tests confirm linear complexity with respect to the number of operators, validating the framework’s feasibility for large-scale, real-time network orchestration.

向6G和开放RAN （O-RAN）的过渡需要跨多运营商网络的智能编排，但这种协作带来了严重的安全和隐私风险。恶意的操作者可能会通过自适应攻击毒害全局模型，而原始梯度的交换则会威胁到数据主权。在本文中，我们提出了FORTRESS-FL，这是一个为安全跨域编排而设计的健壮且保护隐私的联邦学习框架。其核心是TrustChain协议，它协同了一个提交-然后显示的方案，以防止自适应操纵，用于拜占庭检测的无监督谱聚类，以及一个动态声誉系统，以隔离恶意参与者。此外，我们还集成了自适应差分隐私（DP）机制来严格保护操作员数据。对真实世界金融欺诈数据集的广泛评估表明，堡垒-fl对30%拜占庭对手的符号翻转攻击达到100%的检测准确率，防止了在标准基线中观察到的模型分歧。可扩展性测试确认了与运营商数量相关的线性复杂性，验证了该框架在大规模、实时网络编排方面的可行性。

{"title":"FORTRESS-FL: Byzantine-robust and privacy-preserving federated orchestration for next-generation networks","authors":"Quang-Vinh Dang , Thi-Bich-Diem Vo , Ngoc-Son-An Nguyen","doi":"10.1016/j.array.2026.100680","DOIUrl":"10.1016/j.array.2026.100680","url":null,"abstract":"<div><div>The transition to 6G and Open RAN (O-RAN) necessitates intelligent orchestration across multi-operator networks, yet this collaboration introduces severe security and privacy risks. Malicious operators may poison global models through adaptive attacks, while the exchange of raw gradients threatens data sovereignty. In this paper, we propose <strong>FORTRESS-FL</strong>, a robust and privacy-preserving federated learning framework designed for secure cross-domain orchestration. At its core is the <em>TrustChain</em> protocol, which synergizes a commit-then-reveal scheme to prevent adaptive manipulation, unsupervised spectral clustering for Byzantine detection, and a dynamic reputation system to isolate malicious actors. Furthermore, we integrate an adaptive Differential Privacy (DP) mechanism to rigorously protect operator data. Extensive evaluation on a real-world financial fraud dataset demonstrates that FORTRESS-FL achieves 100% detection accuracy against sign-flip attacks with 30% Byzantine adversaries, preventing the model divergence observed in standard baselines. Scalability tests confirm linear complexity with respect to the number of operators, validating the framework’s feasibility for large-scale, real-time network orchestration.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100680"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146184709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel exposine chaotic map for secure image encryption with applications in adaptive cruise control 一种新的用于安全图像加密的暴露混沌映射及其在自适应巡航控制中的应用

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-02-12 DOI: 10.1016/j.array.2026.100710

Malik Obaid Ul Islam , Shahid A. Malik , Pavol Partila , Jaroslav Frnda

Chaotic maps in one dimension often fail to exhibit topological complexity, demonstrating insufficient statistical unpredictability, and deteriorate under finite-precision, producing limited dynamical characteristics, thereby limiting their applicability for lightweight, protected data transmission and control in cost-limited conditions. To address these limitations, we propose a novel one-dimensional Exposine Chaotic Map (ECM) that incorporates sinusoidal-driven variations, exponential reformations, and a dynamic scaling factor to broaden the chaotic band, enhance reactivity to initial conditions, and amplify overall nonlinearity. Moreover, we have also introduced a polynomial approximate version of the proposed ECM using Taylor's and Padé approximations, retaining its chaotic properties while further refining it for utilization in hardware-resource-effective environments. The novel ECM and AECM achieve positive Lyapunov Exponent throughout parameter space, pass all NIST SP 800-22 tests, achieve high correlation dimension, equiform bifurcation, nearly zero autocorrelation, optimum level of entropy, finite-precision adaptability, and robust sensitivity by means of indeterminate, irregular cobweb dynamics. In addition, a Look-Up Table-driven realization strategy is presented to optimize real-time complexity and advance finite-precision interoperability for cost-efficient hardware deployment. Moreover, this work demonstrates the dual applicability of the proposed Exposine map and its approximation in ensuring unpredictability, secure encryption, and cost-efficient dynamic cruise control for intelligent transportation systems. In image cryptosystem, the methodology attains a typical entropy value of 7.9973, NPCR of 99.61%, UACI of 33.47%, a correlation coefficient of 0.0016, key space

2^{302.2862}

, asymptotic intricacy of O(n), and a throughput of 1.0417 MB/s, with runtime of 0.06 s for a 256 × 256 image. In the cruise control implementation, the system acquires an average speed of 60.2915, an approximate entropy of 1.4751, a correlation dimension of 0.1639, a standard deviation of 0.6207, and Shannon entropy of 5.2977, confirming its deterministic chaos, adaptability, and compatibility for real-time adaptive control.

一维混沌映射往往不能表现出拓扑复杂性，表现出统计不可预测性不足，并且在有限精度下恶化，产生有限的动态特性，从而限制了它们在成本有限条件下轻量级、受保护的数据传输和控制的适用性。为了解决这些限制，我们提出了一种新的一维暴露混沌映射（ECM），它包含正弦驱动的变化，指数重构和动态缩放因子，以扩大混沌带，增强对初始条件的反应性，并放大整体非线性。此外，我们还使用Taylor和pad近似引入了所提出的ECM的多项式近似版本，保留了其混沌特性，同时进一步改进其用于硬件资源有效环境。新型ECM和AECM在整个参数空间实现了正Lyapunov指数，通过了NIST SP 800-22的全部测试，通过了不确定的不规则蛛网动力学，实现了高相关维数、等分岔、近零自相关、最优熵水平、有限精度自适应和鲁棒灵敏度。此外，提出了一种查找表驱动的实现策略，以优化实时复杂性和提高有限精度互操作性，从而实现低成本的硬件部署。此外，这项工作证明了所提出的Exposine地图的双重适用性及其在确保智能交通系统的不可预测性、安全加密和成本效益动态巡航控制方面的近似性。在图像密码系统中，该方法的典型熵值为7.9973，NPCR为99.61%，UACI为33.47%，相关系数为0.0016，密钥空间为2302.2862，渐近复杂度为0 (n)，吞吐量为1.0417 MB/s，运行时间为0.06 s。在巡航控制实现中，系统的平均速度为60.2915，近似熵为1.4751，相关维数为0.1639，标准差为0.6207，香农熵为5.2977，证实了系统混沌的确定性、自适应性和实时自适应控制的兼容性。

{"title":"A novel exposine chaotic map for secure image encryption with applications in adaptive cruise control","authors":"Malik Obaid Ul Islam , Shahid A. Malik , Pavol Partila , Jaroslav Frnda","doi":"10.1016/j.array.2026.100710","DOIUrl":"10.1016/j.array.2026.100710","url":null,"abstract":"<div><div>Chaotic maps in one dimension often fail to exhibit topological complexity, demonstrating insufficient statistical unpredictability, and deteriorate under finite-precision, producing limited dynamical characteristics, thereby limiting their applicability for lightweight, protected data transmission and control in cost-limited conditions. To address these limitations, we propose a novel one-dimensional Exposine Chaotic Map (ECM) that incorporates sinusoidal-driven variations, exponential reformations, and a dynamic scaling factor to broaden the chaotic band, enhance reactivity to initial conditions, and amplify overall nonlinearity. Moreover, we have also introduced a polynomial approximate version of the proposed ECM using Taylor's and Padé approximations, retaining its chaotic properties while further refining it for utilization in hardware-resource-effective environments. The novel ECM and AECM achieve positive Lyapunov Exponent throughout parameter space, pass all NIST SP 800-22 tests, achieve high correlation dimension, equiform bifurcation, nearly zero autocorrelation, optimum level of entropy, finite-precision adaptability, and robust sensitivity by means of indeterminate, irregular cobweb dynamics. In addition, a Look-Up Table-driven realization strategy is presented to optimize real-time complexity and advance finite-precision interoperability for cost-efficient hardware deployment. Moreover, this work demonstrates the dual applicability of the proposed Exposine map and its approximation in ensuring unpredictability, secure encryption, and cost-efficient dynamic cruise control for intelligent transportation systems. In image cryptosystem, the methodology attains a typical entropy value of 7.9973, NPCR of 99.61%, UACI of 33.47%, a correlation coefficient of 0.0016, key space <span><math><mrow><msup><mn>2</mn><mn>302.2862</mn></msup></mrow></math></span>, asymptotic intricacy of O(n), and a throughput of 1.0417 MB/s, with runtime of 0.06 s for a 256 × 256 image. In the cruise control implementation, the system acquires an average speed of 60.2915, an approximate entropy of 1.4751, a correlation dimension of 0.1639, a standard deviation of 0.6207, and Shannon entropy of 5.2977, confirming its deterministic chaos, adaptability, and compatibility for real-time adaptive control.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100710"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146184901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI-driven meta-model for cybersecurity in next-generation networks using multilayer Extreme Learning Machine 基于多层极限学习机的下一代网络网络安全的人工智能驱动元模型

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-02-14 DOI: 10.1016/j.array.2026.100707

Jesús Calle-Cancho , Jesús Galeano-Brajones , David Cortés-Polo , Mercedes E. Paoletti , Juan M. Haut

Ensuring robust cybersecurity in next-generation networks has become a critical challenge due to the increasing sophistication of cyber threats, particularly Distributed Denial-of-Service (DDoS) attacks. Traditional detection approaches often suffer from high computational costs, poor adaptability to evolving attack patterns, and limited interpretability. In this work, we propose an AI-driven meta-model, called Meta-Model Multilayer Extreme Learning Machines (MM-MELM), designed to enhance anomaly detection and classification in NGNs. The proposed framework integrates multiple Multilayer Extreme Learning Machine (MELM) models into a meta-learning structure, leveraging the diverse outputs of independently trained MELMs to improve robustness and generalization.

The proposed methodology is evaluated across multiple DDoS attack scenarios, demonstrating its capability to generalize across diverse threat types. Performance analysis reveals that MM-MELM achieves state-of-the-art attack detection, consistently outperforming baseline models. Moreover, MM-MELM exhibits lower variability across all evaluation metrics, ensuring robust performance regardless of the attack complexity. The results highlight that MM-MELM provides a trade-off solution among balanced accuracy, precision, recall, and

F_{1}

-score, making it a highly scalable and adaptive solution for real-time network security.

由于网络威胁日益复杂，特别是分布式拒绝服务（DDoS）攻击，确保下一代网络的强大网络安全已成为一项关键挑战。传统的检测方法通常存在计算成本高、对不断发展的攻击模式适应性差、可解释性有限等问题。在这项工作中，我们提出了一种人工智能驱动的元模型，称为元模型多层极端学习机（MM-MELM），旨在增强ngn中的异常检测和分类。提出的框架将多个多层极限学习机（MELM）模型集成到一个元学习结构中，利用独立训练的MELM的不同输出来提高鲁棒性和泛化。所提出的方法在多个DDoS攻击场景中进行了评估，展示了其在不同威胁类型中泛化的能力。性能分析表明，MM-MELM实现了最先进的攻击检测，始终优于基线模型。此外，MM-MELM在所有评估指标中表现出较低的可变性，无论攻击复杂性如何，都确保了健壮的性能。结果表明，MM-MELM在准确度、精密度、召回率和f1分数之间提供了一种平衡的解决方案，使其成为实时网络安全的高度可扩展和自适应解决方案。

{"title":"AI-driven meta-model for cybersecurity in next-generation networks using multilayer Extreme Learning Machine","authors":"Jesús Calle-Cancho , Jesús Galeano-Brajones , David Cortés-Polo , Mercedes E. Paoletti , Juan M. Haut","doi":"10.1016/j.array.2026.100707","DOIUrl":"10.1016/j.array.2026.100707","url":null,"abstract":"<div><div>Ensuring robust cybersecurity in next-generation networks has become a critical challenge due to the increasing sophistication of cyber threats, particularly Distributed Denial-of-Service (DDoS) attacks. Traditional detection approaches often suffer from high computational costs, poor adaptability to evolving attack patterns, and limited interpretability. In this work, we propose an AI-driven meta-model, called Meta-Model Multilayer Extreme Learning Machines (MM-MELM), designed to enhance anomaly detection and classification in NGNs. The proposed framework integrates multiple Multilayer Extreme Learning Machine (MELM) models into a meta-learning structure, leveraging the diverse outputs of independently trained MELMs to improve robustness and generalization.</div><div>The proposed methodology is evaluated across multiple DDoS attack scenarios, demonstrating its capability to generalize across diverse threat types. Performance analysis reveals that MM-MELM achieves state-of-the-art attack detection, consistently outperforming baseline models. Moreover, MM-MELM exhibits lower variability across all evaluation metrics, ensuring robust performance regardless of the attack complexity. The results highlight that MM-MELM provides a trade-off solution among balanced accuracy, precision, recall, and <span><math><msub><mrow><mi>F</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span>-score, making it a highly scalable and adaptive solution for real-time network security.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100707"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel simulator for performance analysis in heterogeneous wormhole Network-on-Chips 一种用于异构虫洞片上网络性能分析的新型模拟器

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-02-17 DOI: 10.1016/j.array.2026.100715

Md Amirul Islam, Giovanni Stea

This paper introduces scNoCSim (Service Curve-based NoC Simulator), a network-on-chip simulator with integrated monitoring capabilities, built on the omnetpp (OMNeT++) framework. The scNoCSim offers unique features for simulating heterogeneous wormhole Network-on-Chips (NoCs) with monitoring facilities, varying numbers of virtual channels for each unidirectional port, and flexible link capacities. The proposed simulator is extensible and scalable by adding more modules, and highly customizable through editable parameters. It is designed for Network-on-Chip modeling and offers various deterministic routing algorithms (XY, YX, XY–YX, Odd-Even, and O1Turn). It supports round-robin, weighted round-robin, and fixed-priority scheduling strategies; credit-based traffic flow control; and both centralized and distributed routing techniques, including realistic and ideal routers. This simulator offers a comprehensive set of analyses at both packet and flit levels, including, throughput, end-to-end delay, data transfer delay, required bandwidth estimation, link utilization, traffic packet success and loss rates, etc. This paper outlines the proposed architectural view, network design, functional structure, and modules, as well as the key attributes that make it ideal for progressive Network-on-Chip research. Additionally, Network Calculus is used to determine performance (delay and backlog) bounds for worst-case scenarios, estimate bandwidth, and present case studies with validation that demonstrate the capabilities of the scNoCSim simulator in scenarios that cannot be addressed by existing Network-on-Chip research.

本文介绍了基于服务曲线的NoC模拟器scNoCSim (Service Curve-based NoC Simulator)，这是一种基于omnetpp （omnet++）框架的集成监控功能的片上网络模拟器。scNoCSim具有独特的功能，可以模拟具有监控设施的异构虫洞片上网络（noc），为每个单向端口提供不同数量的虚拟通道，以及灵活的链路容量。所提出的模拟器可通过添加更多模块进行扩展和伸缩，并通过可编辑参数进行高度定制。它专为片上网络建模而设计，并提供各种确定性路由算法（XY， YX, XY - YX， Odd-Even和O1Turn）。它支持轮询、加权轮询和固定优先级调度策略；基于信用的交通流量控制；集中式和分布式路由技术，包括现实的和理想的路由器。该模拟器提供了一套全面的分析，在数据包和飞行水平，包括，吞吐量，端到端延迟，数据传输延迟，所需的带宽估计，链路利用率，流量数据包的成功率和损失率等。本文概述了所提出的架构视图、网络设计、功能结构和模块，以及使其成为渐进式片上网络研究的理想的关键属性。此外，网络演算用于确定最坏情况下的性能（延迟和积压）界限，估计带宽，并提供验证案例研究，证明scNoCSim模拟器在现有的片上网络研究无法解决的情况下的能力。

{"title":"A novel simulator for performance analysis in heterogeneous wormhole Network-on-Chips","authors":"Md Amirul Islam, Giovanni Stea","doi":"10.1016/j.array.2026.100715","DOIUrl":"10.1016/j.array.2026.100715","url":null,"abstract":"<div><div>This paper introduces <em>scNoCSim</em> (Service Curve-based NoC Simulator), a network-on-chip simulator with integrated monitoring capabilities, built on the <em>omnetpp</em> (OMNeT++) framework. The <em>scNoCSim</em> offers unique features for simulating heterogeneous wormhole Network-on-Chips (NoCs) with monitoring facilities, varying numbers of virtual channels for each unidirectional port, and flexible link capacities. The proposed simulator is extensible and scalable by adding more modules, and highly customizable through editable parameters. It is designed for Network-on-Chip modeling and offers various deterministic routing algorithms (XY, YX, XY–YX, Odd-Even, and O1Turn). It supports round-robin, weighted round-robin, and fixed-priority scheduling strategies; credit-based traffic flow control; and both centralized and distributed routing techniques, including realistic and ideal routers. This simulator offers a comprehensive set of analyses at both packet and flit levels, including, throughput, end-to-end delay, data transfer delay, required bandwidth estimation, link utilization, traffic packet success and loss rates, etc. This paper outlines the proposed architectural view, network design, functional structure, and modules, as well as the key attributes that make it ideal for progressive Network-on-Chip research. Additionally, Network Calculus is used to determine performance (delay and backlog) bounds for worst-case scenarios, estimate bandwidth, and present case studies with validation that demonstrate the capabilities of the <em>scNoCSim</em> simulator in scenarios that cannot be addressed by existing Network-on-Chip research.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100715"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Breast cancer detection using BCDNet convolutional neural network 利用BCDNet卷积神经网络检测乳腺癌

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-02-18 DOI: 10.1016/j.array.2026.100711

Aneesha Jose, Zoheir Ezziane

Breast cancer detection from histopathological images has benefited significantly from advances in deep learning; however, many high-performing models rely on deep architectures and substantial computational resources, limiting their applicability in resource-constrained clinical settings. Recent studies have demonstrated that convolutional neural networks (CNNs) can effectively support invasive ductal carcinoma (IDC) classification, but often at the cost of increased training time and memory usage [1, 13].

In this study, we propose BCDNet_Updated, an optimized lightweight CNN derived from the reference BCDNet architecture [1] for IDC detection. The proposed model introduces targeted architectural refinements aimed at improving classification performance while preserving computational efficiency. Experiments conducted on the Kaggle IDC dataset demonstrate that BCDNet_Updated achieves improved accuracy, recall, and F1-score compared to the reference BCDNet, while requiring significantly less GPU memory and comparable training time.

While the evaluation is limited to a single dataset and does not include formal statistical significance testing or explainability analysis, the results indicate that careful optimization of lightweight CNN architectures can yield meaningful performance improvements without increasing computational cost. Future work will incorporate multi-dataset validation, statistical rigor through k-fold cross-validation, and explainability mechanisms (Grad-CAM, attention visualization) to support clinical adoption. This work contributes to ongoing efforts toward efficient deep learning models for histopathological image analysis and provides a foundation for future studies involving expanded validation and model interpretability.

从组织病理学图像中检测乳腺癌已经大大受益于深度学习的进步；然而，许多高性能模型依赖于深度架构和大量的计算资源，限制了它们在资源有限的临床环境中的适用性。最近的研究表明，卷积神经网络（cnn）可以有效地支持侵袭性导管癌（invasive ductal carcinoma， IDC）分类，但往往以增加训练时间和内存使用为代价[1,13]。在本研究中，我们提出了BCDNet_Updated，这是一种基于参考BCDNet架构[1]的优化轻量级CNN，用于IDC检测。提出的模型引入了有针对性的架构改进，旨在提高分类性能，同时保持计算效率。在Kaggle IDC数据集上进行的实验表明，与参考BCDNet相比，BCDNet_Updated实现了更高的准确率、召回率和f1分数，同时所需的GPU内存和训练时间显著减少。虽然评估仅限于单个数据集，并且不包括正式的统计显著性测试或可解释性分析，但结果表明，仔细优化轻量级CNN架构可以在不增加计算成本的情况下产生有意义的性能改进。未来的工作将包括多数据集验证，通过k-fold交叉验证的统计严谨性，以及可解释性机制（Grad-CAM，注意力可视化），以支持临床应用。这项工作为组织病理学图像分析的高效深度学习模型的持续努力做出了贡献，并为涉及扩展验证和模型可解释性的未来研究奠定了基础。

{"title":"Breast cancer detection using BCDNet convolutional neural network","authors":"Aneesha Jose, Zoheir Ezziane","doi":"10.1016/j.array.2026.100711","DOIUrl":"10.1016/j.array.2026.100711","url":null,"abstract":"<div><div>Breast cancer detection from histopathological images has benefited significantly from advances in deep learning; however, many high-performing models rely on deep architectures and substantial computational resources, limiting their applicability in resource-constrained clinical settings. Recent studies have demonstrated that convolutional neural networks (CNNs) can effectively support invasive ductal carcinoma (IDC) classification, but often at the cost of increased training time and memory usage [1, 13].</div><div>In this study, we propose BCDNet_Updated, an optimized lightweight CNN derived from the reference BCDNet architecture [1] for IDC detection. The proposed model introduces targeted architectural refinements aimed at improving classification performance while preserving computational efficiency. Experiments conducted on the Kaggle IDC dataset demonstrate that BCDNet_Updated achieves improved accuracy, recall, and F1-score compared to the reference BCDNet, while requiring significantly less GPU memory and comparable training time.</div><div>While the evaluation is limited to a single dataset and does not include formal statistical significance testing or explainability analysis, the results indicate that careful optimization of lightweight CNN architectures can yield meaningful performance improvements without increasing computational cost. Future work will incorporate multi-dataset validation, statistical rigor through k-fold cross-validation, and explainability mechanisms (Grad-CAM, attention visualization) to support clinical adoption. This work contributes to ongoing efforts toward efficient deep learning models for histopathological image analysis and provides a foundation for future studies involving expanded validation and model interpretability.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100711"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Efficient crowd anomaly detection using C3D-LSTM networks with enhanced attention mechanisms 基于增强关注机制的C3D-LSTM网络的高效人群异常检测

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2025-12-03 DOI: 10.1016/j.array.2025.100625

Sarah Altowairqi , Suhuai Luo , Peter Greer , Shan Chen

The rising deployment of surveillance systems in urban environments necessitates efficient automated anomaly detection methods. While showing promise, current deep learning approaches struggle with computational complexity and real-time performance in processing spatiotemporal information. This paper presents a hybrid framework integrating Convolutional 3D Networks (C3D), Long Short-Term Memory (LSTM) networks, and attention mechanisms for anomaly detection. Through a systematic evaluation of four attention mechanisms—self-attention, multi-head attention, Bahdanau attention, and Luong attention—we demonstrate their operational differences and their differential impact on feature extraction and classification performance across three diverse benchmark datasets. Our multi-head attention variant achieves state-of-the-art results with 99.40 % accuracy and 99.96 % Area Under the Curve (AUC) on Violent Flows, while maintaining robust performance across varying dataset complexities, achieving 91.87 % accuracy on the ShanghaiTech Campus and 79.7 % accuracy on the UCF-Crime dataset. Comprehensive cross-dataset evaluation demonstrates consistent improvements of 2.4 %–3.5 % over baseline approaches, with all attention mechanisms outperforming traditional spatiotemporal models. The proposed architecture effectively balances computational requirements with detection performance, maintaining real-time processing capabilities suitable for operational deployment. This framework advances the technical capabilities of anomaly detection systems while providing a validated foundation for practical deployment in diverse surveillance environments, from controlled scenarios to challenging real-world conditions.

越来越多的监控系统部署在城市环境中，需要高效的自动化异常检测方法。虽然显示出前景，但目前的深度学习方法在处理时空信息方面存在计算复杂性和实时性的问题。本文提出了一种将卷积三维网络（C3D）、长短期记忆（LSTM）网络和注意机制集成在一起的混合框架，用于异常检测。通过对四种注意机制（自我注意、多头注意、巴赫达瑙注意和隆注意）的系统评估，我们展示了它们在三个不同基准数据集上的操作差异及其对特征提取和分类性能的差异影响。我们的多头注意力变体在暴力流上实现了最先进的结果，准确率为99.40%，曲线下面积（AUC）为99.96%，同时在不同的数据集复杂性下保持稳健的性能，在上海科技大学校园实现了91.87%的准确率，在ucf -犯罪数据集上实现了79.7%的准确率。综合跨数据集评估表明，与基线方法相比，该方法的一致性改进为2.4% - 3.5%，所有注意力机制都优于传统的时空模型。所提出的体系结构有效地平衡了计算需求和检测性能，保持了适合作战部署的实时处理能力。该框架提高了异常检测系统的技术能力，同时为各种监控环境（从受控场景到具有挑战性的现实条件）的实际部署提供了有效的基础。

{"title":"Efficient crowd anomaly detection using C3D-LSTM networks with enhanced attention mechanisms","authors":"Sarah Altowairqi , Suhuai Luo , Peter Greer , Shan Chen","doi":"10.1016/j.array.2025.100625","DOIUrl":"10.1016/j.array.2025.100625","url":null,"abstract":"<div><div>The rising deployment of surveillance systems in urban environments necessitates efficient automated anomaly detection methods. While showing promise, current deep learning approaches struggle with computational complexity and real-time performance in processing spatiotemporal information. This paper presents a hybrid framework integrating Convolutional 3D Networks (C3D), Long Short-Term Memory (LSTM) networks, and attention mechanisms for anomaly detection. Through a systematic evaluation of four attention mechanisms—self-attention, multi-head attention, Bahdanau attention, and Luong attention—we demonstrate their operational differences and their differential impact on feature extraction and classification performance across three diverse benchmark datasets. Our multi-head attention variant achieves state-of-the-art results with 99.40 % accuracy and 99.96 % Area Under the Curve (AUC) on Violent Flows, while maintaining robust performance across varying dataset complexities, achieving 91.87 % accuracy on the ShanghaiTech Campus and 79.7 % accuracy on the UCF-Crime dataset. Comprehensive cross-dataset evaluation demonstrates consistent improvements of 2.4 %–3.5 % over baseline approaches, with all attention mechanisms outperforming traditional spatiotemporal models. The proposed architecture effectively balances computational requirements with detection performance, maintaining real-time processing capabilities suitable for operational deployment. This framework advances the technical capabilities of anomaly detection systems while providing a validated foundation for practical deployment in diverse surveillance environments, from controlled scenarios to challenging real-world conditions.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100625"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A survey of lightweight methods for object detection networks 目标检测网络的轻量级方法综述

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2025-12-02 DOI: 10.1016/j.array.2025.100589

Jing He, Jianfei Jiang, Changfan Zhang

As social production technologies develop, object detection becomes vital in sectors such as agriculture, industry, and healthcare. It decreases dependence on manual labour and enhances accuracy and efficiency. However, edge devices confront limitations in computational power, storage, and energy, creating a trade-off between accuracy and model size. To tackle this, academia and industry have proposed solutions including hardware-coordinated acceleration, adaptive task lightweighting, and hybrid compression. This paper reviews research from 2020 to 2025 on lightweight object detection, providing a systematic overview of efficient architecture and model compression techniques, explaining their mechanisms, challenges, and future directions to support ongoing progress.

随着社会生产技术的发展，目标检测在农业、工业和医疗保健等部门变得至关重要。它减少了对体力劳动的依赖，提高了准确性和效率。然而，边缘设备面临着计算能力、存储和能量方面的限制，需要在精度和模型尺寸之间进行权衡。为了解决这个问题，学术界和工业界提出了包括硬件协调加速、自适应任务轻量化和混合压缩在内的解决方案。本文回顾了从2020年到2025年在轻量目标检测方面的研究，提供了高效架构和模型压缩技术的系统概述，解释了它们的机制、挑战和未来方向，以支持正在进行的进展。

引用次数: 0

KeepUp: A unified framework fusing knowledge extraction, social platform engagement, and user profiling for fake news detection KeepUp：一个统一的框架，融合了知识提取、社交平台参与和假新闻检测的用户分析

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-01-22 DOI: 10.1016/j.array.2026.100687

Muhammad Wasim , Sehrash Safdar , Abdur Rehman , Zahoor Ur Rehman , Osama A. Khashan , Naif Alzahrani , Anwar Ghani

Approximately half of the global population relies on social media platforms such as Facebook, Twitter, and Instagram for news consumption. The vast volume and rapid dissemination of information on these platforms pose substantial challenges for the timely and accurate detection of fake news. Academics are working harder to develop AI-based automated systems to check news accuracy because of the detrimental effects of misinformation on public health, social trust, and political stability. But the majority of false news detection methods currently in use focus primarily on content-based features, often ignoring essential factors such as user profiling, social context, and knowledge extraction. The knowledge-based features necessary for effective document retrieval, position identification, social engagement analysis, and user profile integration are often absent from datasets, even though some of them contain elements of social context and user behavior. This work offers a thorough, fully annotated dataset that integrates user profiles, stance information, social engagements, knowledge extraction, and content elements into a single resource to overcome these limitations. Building on this dataset, this study creates KeepUp, a unified system that integrates user profiles, social media activity, and knowledge extraction to detect bogus news. KeepUp outperforms all baseline models, achieving a detection accuracy of 0.78, demonstrating the effectiveness of this combined approach.

全球大约一半的人口依赖Facebook、Twitter和Instagram等社交媒体平台来消费新闻。这些平台上信息的庞大数量和快速传播对及时准确地发现假新闻构成了重大挑战。学者们正在努力开发基于人工智能的自动化系统，以检查新闻的准确性，因为错误信息对公共健康、社会信任和政治稳定产生了有害影响。但是，目前使用的大多数假新闻检测方法主要集中在基于内容的特征上，往往忽略了用户特征、社会背景和知识提取等基本因素。有效的文档检索、职位识别、社会参与分析和用户档案集成所必需的基于知识的特征通常不在数据集中，尽管其中一些数据集包含社会背景和用户行为的元素。这项工作提供了一个全面的、完全注释的数据集，它将用户配置文件、立场信息、社交活动、知识提取和内容元素集成到一个资源中，以克服这些限制。在此数据集的基础上，本研究创建了KeepUp，这是一个统一的系统，集成了用户配置文件、社交媒体活动和知识提取来检测虚假新闻。KeepUp优于所有基线模型，达到0.78的检测精度，证明了这种组合方法的有效性。

{"title":"KeepUp: A unified framework fusing knowledge extraction, social platform engagement, and user profiling for fake news detection","authors":"Muhammad Wasim , Sehrash Safdar , Abdur Rehman , Zahoor Ur Rehman , Osama A. Khashan , Naif Alzahrani , Anwar Ghani","doi":"10.1016/j.array.2026.100687","DOIUrl":"10.1016/j.array.2026.100687","url":null,"abstract":"<div><div>Approximately half of the global population relies on social media platforms such as Facebook, Twitter, and Instagram for news consumption. The vast volume and rapid dissemination of information on these platforms pose substantial challenges for the timely and accurate detection of fake news. Academics are working harder to develop AI-based automated systems to check news accuracy because of the detrimental effects of misinformation on public health, social trust, and political stability. But the majority of false news detection methods currently in use focus primarily on content-based features, often ignoring essential factors such as user profiling, social context, and knowledge extraction. The knowledge-based features necessary for effective document retrieval, position identification, social engagement analysis, and user profile integration are often absent from datasets, even though some of them contain elements of social context and user behavior. This work offers a thorough, fully annotated dataset that integrates user profiles, stance information, social engagements, knowledge extraction, and content elements into a single resource to overcome these limitations. Building on this dataset, this study creates KeepUp, a unified system that integrates user profiles, social media activity, and knowledge extraction to detect bogus news. KeepUp outperforms all baseline models, achieving a detection accuracy of 0.78, demonstrating the effectiveness of this combined approach.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100687"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI-clinic fusion: Multimodal deep learning redefines IBD diagnostics via synergistic CT-clinical feature integration ai -临床融合：多模式深度学习通过协同的ct -临床特征整合重新定义IBD诊断

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-03-01 Epub Date: 2026-02-02 DOI: 10.1016/j.array.2026.100702

Yong Huang , Pan Li , Xiao-N Zhong , DongPu Luo , Yin Zhou , Yong-J Li , Ming-hong Sun , Li-J Zhang , Li-H Yang , JunLong Li , Yi Tao

Background

Traditional methods like endoscopy and imaging are important for IBD management, which are limited by operator dependency, complexity, and high costs, making standardization and routine monitoring difficult.

Methods

In this study, seven machine learning algorithms were applied to the UK Biobank database to optimize the selection of clinical features and identify key variables associated with IBD. The nnUNetv2 model was then used for CT gut segmentation and validated by a local dataset. Next, the YOLOv10 algorithm was employed to accurately detect IBD lesion regions in the intestine. Finally, an IBD diagnostic model was developed by integrating the selected clinical features with the YOLOv10 identification results, thereby demonstrating the advantages of multimodal data integration.

Results

A machine learning-driven feature selection identified chronic abdominal pain, LDL-C, gender, TC, RDW, WBC, ALB, HDL-C, and neutrophil counts as clinical predictors. The nnUNetv2 model achieved high-precision intestinal segmentation, while YOLOv10 demonstrated robust lesion detection. A multimodal nomogram combining clinical features with YOLOv10-derived imaging biomarkers significantly enhanced IBD diagnosis, achieving an AUC of 0.967 that surpassed the clinical-only model (AUC = 0.730). This improvement highlights the synergistic value of deep learning-augmented diagnostics for classification accuracy.

Conclusion

Integrating clinical features with AI-driven image analysis improves IBD diagnostic precision, enabling better clinical decision-making and patient outcomes.

Novelty

To our knowledge, this is the first multimodal nomogram that combines the predictive clinical features of the UK Biobank scale with CT-based YOLOv10 lesion detection for IBD diagnosis. Different from prior models that solely rely on clinical data or radiomics, our approach achieves integration of population-level epidemiology and deep learning-based imaging, and was validated through external datasets.

内镜检查和影像学等传统方法对IBD的治疗非常重要，但受操作者依赖性、复杂性和高成本的限制，使得标准化和常规监测变得困难。方法本研究将7种机器学习算法应用于UK Biobank数据库，优化临床特征的选择，识别与IBD相关的关键变量。然后将nnUNetv2模型用于CT肠道分割，并通过本地数据集进行验证。接下来，使用YOLOv10算法准确检测肠内IBD病变区域。最后，将选择的临床特征与YOLOv10识别结果相结合，建立IBD诊断模型，从而展示了多模式数据集成的优势。结果机器学习驱动的特征选择将慢性腹痛、LDL-C、性别、TC、RDW、WBC、ALB、HDL-C和中性粒细胞计数确定为临床预测因子。nnUNetv2模型实现了高精度的肠道分割，而YOLOv10模型显示了强大的病变检测能力。结合临床特征和yolov10衍生的成像生物标志物的多模态图显著提高了IBD的诊断，AUC达到0.967，超过了仅临床模型（AUC = 0.730）。这一改进突出了深度学习增强诊断对分类准确性的协同价值。结论将临床特征与人工智能驱动的图像分析相结合，提高了IBD的诊断精度，有助于更好的临床决策和患者预后。据我们所知，这是第一个将UK Biobank量表的预测临床特征与基于ct的YOLOv10病变检测相结合用于IBD诊断的多模态图。与以往仅依赖临床数据或放射组学的模型不同，我们的方法实现了人口水平流行病学和基于深度学习的成像的整合，并通过外部数据集进行了验证。

{"title":"AI-clinic fusion: Multimodal deep learning redefines IBD diagnostics via synergistic CT-clinical feature integration","authors":"Yong Huang , Pan Li , Xiao-N Zhong , DongPu Luo , Yin Zhou , Yong-J Li , Ming-hong Sun , Li-J Zhang , Li-H Yang , JunLong Li , Yi Tao","doi":"10.1016/j.array.2026.100702","DOIUrl":"10.1016/j.array.2026.100702","url":null,"abstract":"<div><h3>Background</h3><div>Traditional methods like endoscopy and imaging are important for IBD management, which are limited by operator dependency, complexity, and high costs, making standardization and routine monitoring difficult.</div></div><div><h3>Methods</h3><div>In this study, seven machine learning algorithms were applied to the UK Biobank database to optimize the selection of clinical features and identify key variables associated with IBD. The nnUNetv2 model was then used for CT gut segmentation and validated by a local dataset. Next, the YOLOv10 algorithm was employed to accurately detect IBD lesion regions in the intestine. Finally, an IBD diagnostic model was developed by integrating the selected clinical features with the YOLOv10 identification results, thereby demonstrating the advantages of multimodal data integration.</div></div><div><h3>Results</h3><div>A machine learning-driven feature selection identified chronic abdominal pain, LDL-C, gender, TC, RDW, WBC, ALB, HDL-C, and neutrophil counts as clinical predictors. The nnUNetv2 model achieved high-precision intestinal segmentation, while YOLOv10 demonstrated robust lesion detection. A multimodal nomogram combining clinical features with YOLOv10-derived imaging biomarkers significantly enhanced IBD diagnosis, achieving an AUC of 0.967 that surpassed the clinical-only model (AUC = 0.730). This improvement highlights the synergistic value of deep learning-augmented diagnostics for classification accuracy.</div></div><div><h3>Conclusion</h3><div>Integrating clinical features with AI-driven image analysis improves IBD diagnostic precision, enabling better clinical decision-making and patient outcomes.</div></div><div><h3>Novelty</h3><div>To our knowledge, this is the first multimodal nomogram that combines the predictive clinical features of the UK Biobank scale with CT-based YOLOv10 lesion detection for IBD diagnosis. Different from prior models that solely rely on clinical data or radiomics, our approach achieves integration of population-level epidemiology and deep learning-based imaging, and was validated through external datasets.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100702"},"PeriodicalIF":4.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146184685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Array最新文献

Background

Methods

Results

Conclusion

Novelty