Journal of King Saud University-Computer and Information Sciences最新文献_第2页

General secure encryption algorithm for separable reversible data hiding in encrypted domain 加密域中可分离可逆数据隐藏的通用安全加密算法

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-11-01 DOI: 10.1016/j.jksuci.2024.102217

Hongli Wan, Minqing Zhang, Yan Ke, Zongbao Jiang, Fuqiang Di

The separable reversible data hiding in encrypted domain (RDH-ED) algorithm leaves out the embedding space for the information before or after encryption and makes the operation of extracting the information and restoring the image not interfere with each other. The encryption method employed not only affects the embedding space of the information and separability, but is more crucial for ensuring security. However, the commonly used XOR, scram-bling or combination methods fall short in security, especially against known plaintext attack (KPA). Therefore, in order to improve the security of RDH-ED and be widely applicable, this paper proposes a high-security RDH-ED encryption algorithm that can be used to reserve space before encryption (RSBE) and free space after encryption (FSAE). During encryption, the image undergoes block XOR, global intra-block bit-plane scrambling (GIBS) and inter-block scrambling sequentially. The GIBS key is created through chaotic mapping transformation. Subsequently, two RDH-ED algorithms based on this encryption are proposed. Experimental results indicate that the algorithm outlined in this paper maintains consistent key communication traffic post key conversion. Additionally, its computational complexity remains at a constant level, satisfying separability criteria, and is suitable for both RSBE and FSAE methods. Simultaneously, while satisfying the security of a single encryption technique, we have expanded the key space to 2

^{8 N p} \times N p! \times 8!^{N p}

, enabling resilience against various existing attack methods. Notably, particularly in KPA testing scenarios, the average decryption success rate is a mere 0.0067% and 0.0045%, highlighting its exceptional security. Overall, this virtually unbreakable system significantly enhances image security while preserving an appropriate embedding capacity.

加密域中的可分离可逆数据隐藏（RDH-ED）算法在加密前后都留出了信息的嵌入空间，使提取信息和还原图像的操作互不干扰。所采用的加密方法不仅会影响信息的嵌入空间和可分离性，而且对确保安全性更为关键。然而，常用的 XOR、加扰或组合方法在安全性方面存在不足，尤其是在应对已知明文攻击（KPA）时。因此，为了提高 RDH-ED 的安全性和广泛适用性，本文提出了一种可用于加密前预留空间（RSBE）和加密后释放空间（FSAE）的高安全性 RDH-ED 加密算法。在加密过程中，图像依次经过块 XOR、全局块内位平面加扰（GIBS）和块间加扰。GIBS 密钥通过混沌映射变换创建。随后，提出了两种基于这种加密的 RDH-ED 算法。实验结果表明，本文概述的算法能在密钥转换后保持一致的密钥通信流量。此外，该算法的计算复杂度保持在恒定水平，满足可分性标准，同时适用于 RSBE 和 FSAE 方法。同时，在满足单一加密技术安全性的同时，我们还将密钥空间扩展到了 28Np×Np！×8！Np，从而能够抵御现有的各种攻击方法。值得注意的是，特别是在 KPA 测试场景中，平均解密成功率仅为 0.0067% 和 0.0045%，彰显了其卓越的安全性。总之，这个几乎牢不可破的系统在保持适当嵌入容量的同时，大大增强了图像的安全性。

{"title":"General secure encryption algorithm for separable reversible data hiding in encrypted domain","authors":"Hongli Wan, Minqing Zhang, Yan Ke, Zongbao Jiang, Fuqiang Di","doi":"10.1016/j.jksuci.2024.102217","DOIUrl":"10.1016/j.jksuci.2024.102217","url":null,"abstract":"<div><div>The separable reversible data hiding in encrypted domain (RDH-ED) algorithm leaves out the embedding space for the information before or after encryption and makes the operation of extracting the information and restoring the image not interfere with each other. The encryption method employed not only affects the embedding space of the information and separability, but is more crucial for ensuring security. However, the commonly used XOR, scram-bling or combination methods fall short in security, especially against known plaintext attack (KPA). Therefore, in order to improve the security of RDH-ED and be widely applicable, this paper proposes a high-security RDH-ED encryption algorithm that can be used to reserve space before encryption (RSBE) and free space after encryption (FSAE). During encryption, the image undergoes block XOR, global intra-block bit-plane scrambling (GIBS) and inter-block scrambling sequentially. The GIBS key is created through chaotic mapping transformation. Subsequently, two RDH-ED algorithms based on this encryption are proposed. Experimental results indicate that the algorithm outlined in this paper maintains consistent key communication traffic post key conversion. Additionally, its computational complexity remains at a constant level, satisfying separability criteria, and is suitable for both RSBE and FSAE methods. Simultaneously, while satisfying the security of a single encryption technique, we have expanded the key space to 2<span><math><mrow><msup><mrow></mrow><mrow><mn>8</mn><mi>N</mi><mi>p</mi></mrow></msup><mo>×</mo><mi>N</mi><mi>p</mi><mo>!</mo><mo>×</mo><mn>8</mn><msup><mrow><mo>!</mo></mrow><mrow><mi>N</mi><mi>p</mi></mrow></msup></mrow></math></span>, enabling resilience against various existing attack methods. Notably, particularly in KPA testing scenarios, the average decryption success rate is a mere 0.0067% and 0.0045%, highlighting its exceptional security. Overall, this virtually unbreakable system significantly enhances image security while preserving an appropriate embedding capacity.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102217"},"PeriodicalIF":5.2,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142578586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Quantum computing enhanced knowledge tracing: Personalized KT research for mitigating data sparsity 量子计算增强知识追踪：缓解数据稀疏性的个性化 KT 研究

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-11-01 DOI: 10.1016/j.jksuci.2024.102224

Chengke Bao , Qianxi Wu , Weidong Ji , Min Wang , Haoyu Wang

With the development of artificial intelligence in education, knowledge tracing (KT) has become a current research hotspot and is the key to the success of personalized instruction. However, data sparsity remains a significant challenge in the KT domain. To address this challenge, this paper applies quantum computing (QC) technology to KT for the first time. It proposes two personalized KT models incorporating quantum mechanics (QM): quantum convolutional enhanced knowledge tracing (QCE-KT) and quantum variational enhanced knowledge tracing (QVE-KT). Through quantum superposition and entanglement properties, QCE-KT and QVE-KT effectively alleviate the data sparsity problem in the KT domain through quantum convolutional layers and variational quantum circuits, respectively, and significantly improve the quality of the representation and prediction accuracy of students’ knowledge states. Experiments on three datasets show that our models outperform ten benchmark models. On the most sparse dataset, QCE-KT and QVE-KT improve their performance by 16.44% and 14.78%, respectively, compared to DKT. Although QC is still in the developmental stage, this study reveals the great potential of QM in personalized KT, which provides new perspectives for solving personalized instruction problems and opens up new directions for applying QC in education.

随着人工智能在教育领域的发展，知识追踪（KT）已成为当前的研究热点，也是个性化教学成功的关键。然而，数据稀疏性仍然是知识追踪领域的一个重大挑战。为应对这一挑战，本文首次将量子计算（QC）技术应用于 KT。它提出了两种结合量子力学（QM）的个性化知识追踪模型：量子卷积增强知识追踪（QCE-KT）和量子变分增强知识追踪（QVE-KT）。通过量子叠加和纠缠特性，QCE-KT 和 QVE-KT 分别通过量子卷积层和量子变分电路有效缓解了知识追踪领域的数据稀疏性问题，显著提高了学生知识状态的表征质量和预测精度。三个数据集的实验表明，我们的模型优于十个基准模型。在最稀疏的数据集上，QCE-KT 和 QVE-KT 的性能比 DKT 分别提高了 16.44% 和 14.78%。虽然 QC 仍处于发展阶段，但本研究揭示了 QM 在个性化 KT 中的巨大潜力，为解决个性化教学问题提供了新的视角，也为 QC 在教育领域的应用开辟了新的方向。

{"title":"Quantum computing enhanced knowledge tracing: Personalized KT research for mitigating data sparsity","authors":"Chengke Bao , Qianxi Wu , Weidong Ji , Min Wang , Haoyu Wang","doi":"10.1016/j.jksuci.2024.102224","DOIUrl":"10.1016/j.jksuci.2024.102224","url":null,"abstract":"<div><div>With the development of artificial intelligence in education, knowledge tracing (KT) has become a current research hotspot and is the key to the success of personalized instruction. However, data sparsity remains a significant challenge in the KT domain. To address this challenge, this paper applies quantum computing (QC) technology to KT for the first time. It proposes two personalized KT models incorporating quantum mechanics (QM): quantum convolutional enhanced knowledge tracing (QCE-KT) and quantum variational enhanced knowledge tracing (QVE-KT). Through quantum superposition and entanglement properties, QCE-KT and QVE-KT effectively alleviate the data sparsity problem in the KT domain through quantum convolutional layers and variational quantum circuits, respectively, and significantly improve the quality of the representation and prediction accuracy of students’ knowledge states. Experiments on three datasets show that our models outperform ten benchmark models. On the most sparse dataset, QCE-KT and QVE-KT improve their performance by 16.44% and 14.78%, respectively, compared to DKT. Although QC is still in the developmental stage, this study reveals the great potential of QM in personalized KT, which provides new perspectives for solving personalized instruction problems and opens up new directions for applying QC in education.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102224"},"PeriodicalIF":5.2,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DA-Net: A classification-guided network for dental anomaly detection from dental and maxillofacial images DA-Net：从牙科和颌面部图像中检测牙科异常的分类指导网络

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-11-01 DOI: 10.1016/j.jksuci.2024.102229

Jiaxing Li

Dental abnormalities (DA) are frequent signs of disorders of the mouth that cause discomfort, infection, and loss of teeth. Early and reasonably priced treatment may be possible if defective teeth in the oral cavity are automatically detected. Several research works have endeavored to create a potent deep learning model capable of identifying DA from pictures. However, because of the following problems, aberrant teeth from the oral cavity are difficult to detect: 1) Normal teeth and crowded dentition frequently overlap; 2) The lesion area on the tooth surface is tiny. This paper proposes a professional dental anomaly detection network (DA-Net) to address such issues. First, a multi-scale dense connection module (MSDC) is designed to distinguish crowded teeth from normal teeth by learning multi-scale spatial information of dentition. Then, a pixel differential convolution (PDC) module is designed to perform pathological tooth recognition by extracting small lesion features. Finally, a multi-stage convolutional attention module (MSCA) is developed to integrate spatial information and channel information to obtain abnormal teeth in small areas. Experiments on benchmarks show that DA-Net performs well in dental anomaly detection and can further assist doctors in making treatment plans. Specifically, the DA-Net method performs best on multiple detection evaluation metrics: IoU, PRE, REC, and mAP. In terms of REC and mAP indicators, the proposed DA-Net method is 1.1% and 1.3% higher than the second-ranked YOLOv7 method.

牙齿异常（DA）是口腔疾病的常见征兆，会引起不适、感染和牙齿脱落。如果能自动检测出口腔中存在缺陷的牙齿，就可以及早进行价格合理的治疗。一些研究工作致力于创建一个强大的深度学习模型，能够从图片中识别牙齿缺损。然而，由于以下问题，口腔畸形牙难以检测：1）正常牙齿和拥挤牙经常重叠；2）牙齿表面的病变面积很小。针对这些问题，本文提出了一种专业的牙齿异常检测网络（DA-Net）。首先，设计了一个多尺度密集连接模块（MSDC），通过学习牙列的多尺度空间信息来区分拥挤牙和正常牙。然后，设计了一个像素差分卷积（PDC）模块，通过提取小病变特征来进行病牙识别。最后，开发了多级卷积注意力模块（MSCA），以整合空间信息和通道信息，从而获得小区域的异常牙齿。基准实验表明，DA-Net 在牙齿异常检测方面表现出色，可以进一步帮助医生制定治疗方案。具体来说，DA-Net 方法在多个检测评估指标上表现最佳：IoU、PRE、REC 和 mAP。在 REC 和 mAP 指标上，DA-Net 方法比排名第二的 YOLOv7 方法分别高出 1.1% 和 1.3%。

{"title":"DA-Net: A classification-guided network for dental anomaly detection from dental and maxillofacial images","authors":"Jiaxing Li","doi":"10.1016/j.jksuci.2024.102229","DOIUrl":"10.1016/j.jksuci.2024.102229","url":null,"abstract":"<div><div>Dental abnormalities (DA) are frequent signs of disorders of the mouth that cause discomfort, infection, and loss of teeth. Early and reasonably priced treatment may be possible if defective teeth in the oral cavity are automatically detected. Several research works have endeavored to create a potent deep learning model capable of identifying DA from pictures. However, because of the following problems, aberrant teeth from the oral cavity are difficult to detect: 1) Normal teeth and crowded dentition frequently overlap; 2) The lesion area on the tooth surface is tiny. This paper proposes a professional dental anomaly detection network (DA-Net) to address such issues. First, a multi-scale dense connection module (MSDC) is designed to distinguish crowded teeth from normal teeth by learning multi-scale spatial information of dentition. Then, a pixel differential convolution (PDC) module is designed to perform pathological tooth recognition by extracting small lesion features. Finally, a multi-stage convolutional attention module (MSCA) is developed to integrate spatial information and channel information to obtain abnormal teeth in small areas. Experiments on benchmarks show that DA-Net performs well in dental anomaly detection and can further assist doctors in making treatment plans. Specifically, the DA-Net method performs best on multiple detection evaluation metrics: IoU, PRE, REC, and mAP. In terms of REC and mAP indicators, the proposed DA-Net method is 1.1% and 1.3% higher than the second-ranked YOLOv7 method.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102229"},"PeriodicalIF":5.2,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142578584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep indoor illumination estimation based on spherical gaussian representation with scene prior knowledge 基于球形高斯表示和场景先验知识的深度室内光照度估计

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-23 DOI: 10.1016/j.jksuci.2024.102222

Chao Xu , Cheng Han , Huamin Yang , Chao Zhang , Shiyu Lu

High dynamic range (HDR) illumination estimation from a single low dynamic range image is a critical task in the fields of computer vision, graphics and augmented reality. However, directly learning the full HDR environment map or parametric lighting information from a single image is extremely difficult and inaccurate. As a result, we propose a two-stage network approach for illumination estimation that integrates spherical gaussian (SG) representation with scene prior knowledge. In the first stage, a convolutional neural network is utilized to generate material and geometric information about the scene, which serves as prior knowledge for lighting prediction. In the second stage, we model indoor environment illumination using 128 SG functions with fixed center direction and bandwidth, allowing only the amplitude to vary. Subsequently, a Transformer-based lighting parameter regressor is employed to capture the complex relationship between the input images with scene prior information and its SG illumination. Additionally, we introduce a hybrid loss function, which combines a masked loss for high-frequency illumination with a rendering loss for improving the visual quality. By training and evaluating the lighting model on the created SG illumination dataset, the proposed method achieves competitive results in both quantitative metrics and visual quality, outperforming state-of-the-art methods.

从单张低动态范围图像估算高动态范围（HDR）照明是计算机视觉、图形学和增强现实领域的一项重要任务。然而，直接从单张图像中学习完整的 HDR 环境图或参数照明信息是极其困难和不准确的。因此，我们提出了一种将球形高斯（SG）表示法与场景先验知识相结合的两阶段光照估计网络方法。在第一阶段，利用卷积神经网络生成有关场景的材料和几何信息，作为照明预测的先验知识。在第二阶段，我们使用 128 个具有固定中心方向和带宽的 SG 函数对室内环境照明进行建模，只允许振幅变化。随后，我们采用了基于变压器的照明参数回归器，以捕捉输入图像与场景先验信息及其 SG 照明之间的复杂关系。此外，我们还引入了一种混合损失函数，它结合了用于高频照明的遮蔽损失和用于改善视觉质量的渲染损失。通过在创建的 SG 照明数据集上训练和评估照明模型，所提出的方法在定量指标和视觉质量方面都取得了有竞争力的结果，优于最先进的方法。

{"title":"Deep indoor illumination estimation based on spherical gaussian representation with scene prior knowledge","authors":"Chao Xu , Cheng Han , Huamin Yang , Chao Zhang , Shiyu Lu","doi":"10.1016/j.jksuci.2024.102222","DOIUrl":"10.1016/j.jksuci.2024.102222","url":null,"abstract":"<div><div>High dynamic range (HDR) illumination estimation from a single low dynamic range image is a critical task in the fields of computer vision, graphics and augmented reality. However, directly learning the full HDR environment map or parametric lighting information from a single image is extremely difficult and inaccurate. As a result, we propose a two-stage network approach for illumination estimation that integrates spherical gaussian (SG) representation with scene prior knowledge. In the first stage, a convolutional neural network is utilized to generate material and geometric information about the scene, which serves as prior knowledge for lighting prediction. In the second stage, we model indoor environment illumination using 128 SG functions with fixed center direction and bandwidth, allowing only the amplitude to vary. Subsequently, a Transformer-based lighting parameter regressor is employed to capture the complex relationship between the input images with scene prior information and its SG illumination. Additionally, we introduce a hybrid loss function, which combines a masked loss for high-frequency illumination with a rendering loss for improving the visual quality. By training and evaluating the lighting model on the created SG illumination dataset, the proposed method achieves competitive results in both quantitative metrics and visual quality, outperforming state-of-the-art methods.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 10","pages":"Article 102222"},"PeriodicalIF":5.2,"publicationDate":"2024-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142657924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhanced UrduAspectNet: Leveraging Biaffine Attention for superior Aspect-Based Sentiment Analysis 增强型 UrduAspectNet：利用双峰注意力实现卓越的基于方面的情感分析

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-23 DOI: 10.1016/j.jksuci.2024.102221

Kamran Aziz , Naveed Ahmed , Hassan Jalil Hadi , Aizihaierjiang Yusufu , Mohammaed Ali Alshara , Yasir Javed , Donghong Ji

Urdu, with its rich linguistic complexity, poses significant challenges for computational sentiment analysis. This study presents an enhanced version of UrduAspectNet, specifically designed for Aspect-Based Sentiment Analysis (ABSA) in Urdu. We introduce key innovations including the incorporation of Biaffine Attention into the model architecture, which synergizes XLM-R embeddings, a bidirectional LSTM (BiLSTM), and dual Graph Convolutional Networks (GCNs). Additionally, we utilize dependency parsing to create the adjacency matrix for the GCNs, capturing syntactic dependencies to enhance relational representation. The improved model, termed Enhanced UrduAspectNet, integrates POS and lemma embeddings, processed through BiLSTM and GCN layers, with Biaffine Attention enhancing the extraction of intricate aspect and sentiment relationships. We also introduce the use of BIO tags for aspect term identification, improving the granularity of aspect extraction. Experimental results demonstrate significant improvements in both aspect extraction and sentiment classification accuracy. This research advances Urdu sentiment analysis and sets a precedent for leveraging sophisticated NLP techniques in underrepresented languages.

乌尔都语具有丰富的语言复杂性，给计算情感分析带来了巨大挑战。本研究介绍了 UrduAspectNet 的增强版，该版本专为基于方面的乌尔都语情感分析 (ABSA) 而设计。我们引入了一些关键的创新，包括在模型架构中加入 Biaffine Attention，使 XLM-R 嵌入、双向 LSTM（BiLSTM）和双图卷积网络（GCN）协同增效。此外，我们还利用依赖性解析为 GCNs 创建邻接矩阵，捕捉句法依赖性以增强关系表示。改进后的模型被称为 "增强型 UrduAspectNet"，它将通过 BiLSTM 和 GCN 层处理的 POS 和词素嵌入与 Biaffine Attention 整合在一起，从而增强了对错综复杂的方面和情感关系的提取。我们还引入了 BIO 标签用于方面术语识别，从而提高了方面提取的粒度。实验结果表明，方面提取和情感分类的准确性都有显著提高。这项研究推动了乌尔都语情感分析的发展，为在代表性不足的语言中利用复杂的 NLP 技术开创了先例。

{"title":"Enhanced UrduAspectNet: Leveraging Biaffine Attention for superior Aspect-Based Sentiment Analysis","authors":"Kamran Aziz , Naveed Ahmed , Hassan Jalil Hadi , Aizihaierjiang Yusufu , Mohammaed Ali Alshara , Yasir Javed , Donghong Ji","doi":"10.1016/j.jksuci.2024.102221","DOIUrl":"10.1016/j.jksuci.2024.102221","url":null,"abstract":"<div><div>Urdu, with its rich linguistic complexity, poses significant challenges for computational sentiment analysis. This study presents an enhanced version of UrduAspectNet, specifically designed for Aspect-Based Sentiment Analysis (ABSA) in Urdu. We introduce key innovations including the incorporation of Biaffine Attention into the model architecture, which synergizes XLM-R embeddings, a bidirectional LSTM (BiLSTM), and dual Graph Convolutional Networks (GCNs). Additionally, we utilize dependency parsing to create the adjacency matrix for the GCNs, capturing syntactic dependencies to enhance relational representation. The improved model, termed Enhanced UrduAspectNet, integrates POS and lemma embeddings, processed through BiLSTM and GCN layers, with Biaffine Attention enhancing the extraction of intricate aspect and sentiment relationships. We also introduce the use of BIO tags for aspect term identification, improving the granularity of aspect extraction. Experimental results demonstrate significant improvements in both aspect extraction and sentiment classification accuracy. This research advances Urdu sentiment analysis and sets a precedent for leveraging sophisticated NLP techniques in underrepresented languages.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102221"},"PeriodicalIF":5.2,"publicationDate":"2024-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142529770","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Echocardiographic mitral valve segmentation model 超声心动图二尖瓣分割模型

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-19 DOI: 10.1016/j.jksuci.2024.102218

Chunxia Liu , Shanshan Dong , Feng Xiong , Luqing Wang , Bolun Li , Hongjun Wang

Segmentation of mitral valve is not only important for clinical diagnosis, but also has far-reaching impact on prevention and prognosis of the disease by experts and doctors. In this paper, the multi-channel cross fusion transformer based U-Net network model (MCCT-UNet) is proposed according to the classical U-Net architecture. First, the jump connection part of MCCT-UNet is designed by using a multi-channel cross-fusion based attention mechanism module (MCCT) instead of the original jump connection, and this module fuses the feature maps from different scales in different stages of the encoder. Second, the optimization of the feature fusion method is proposed in the decoding stage by designing the cross-compression excitation sub-module (C-SENet) to replace the simple feature splicing, and the C-SENet is used to bridge the inconsistency of the semantic hierarchy by effectively combining the deeper information in the encoding stage with the shallower information. This two modules can establish a close connection between the encoder and decoder by exploring multi-scale global contextual information to solve the semantic divide problem, thus it significantly improves the segmentation performance of the network. The experimental results show that the improvement is effective, and the MCCT-UNet model outperforms the other 9 network models. Specifically, the MCCT-UNet achieved a Dice coefficient of 0.8734, an IoU of 0.7854, and an accuracy of 0.9977, demonstrating significant improvements over the compared models.

二尖瓣的分割不仅对临床诊断有重要意义，而且对专家和医生预防和预后疾病也有深远影响。本文根据经典的 U-Net 架构，提出了基于多通道交叉融合变压器的 U-Net 网络模型（MCCT-UNet）。首先，MCCT-UNet 的跳转连接部分采用基于多通道交叉融合的注意力机制模块（MCCT）代替原有的跳转连接，该模块在编码器的不同阶段融合不同尺度的特征图。其次，在解码阶段提出了对特征融合方法的优化，设计了交叉压缩激发子模块（C-SENet）来替代简单的特征拼接，通过 C-SENet 将编码阶段的深层信息与浅层信息有效结合，弥合语义层次的不一致性。这两个模块通过探索多尺度的全局上下文信息，在编码器和解码器之间建立了紧密的联系，从而解决了语义鸿沟问题，显著提高了网络的分割性能。实验结果表明，改进效果显著，MCCT-UNet 模型优于其他 9 个网络模型。具体来说，MCCT-UNet 的骰子系数达到了 0.8734，IoU 达到了 0.7854，准确率达到了 0.9977，与其他模型相比有了显著的提高。

{"title":"Echocardiographic mitral valve segmentation model","authors":"Chunxia Liu , Shanshan Dong , Feng Xiong , Luqing Wang , Bolun Li , Hongjun Wang","doi":"10.1016/j.jksuci.2024.102218","DOIUrl":"10.1016/j.jksuci.2024.102218","url":null,"abstract":"<div><div>Segmentation of mitral valve is not only important for clinical diagnosis, but also has far-reaching impact on prevention and prognosis of the disease by experts and doctors. In this paper, the multi-channel cross fusion transformer based U-Net network model (MCCT-UNet) is proposed according to the classical U-Net architecture. First, the jump connection part of MCCT-UNet is designed by using a multi-channel cross-fusion based attention mechanism module (MCCT) instead of the original jump connection, and this module fuses the feature maps from different scales in different stages of the encoder. Second, the optimization of the feature fusion method is proposed in the decoding stage by designing the cross-compression excitation sub-module (C-SENet) to replace the simple feature splicing, and the C-SENet is used to bridge the inconsistency of the semantic hierarchy by effectively combining the deeper information in the encoding stage with the shallower information. This two modules can establish a close connection between the encoder and decoder by exploring multi-scale global contextual information to solve the semantic divide problem, thus it significantly improves the segmentation performance of the network. The experimental results show that the improvement is effective, and the MCCT-UNet model outperforms the other 9 network models. Specifically, the MCCT-UNet achieved a Dice coefficient of 0.8734, an IoU of 0.7854, and an accuracy of 0.9977, demonstrating significant improvements over the compared models.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102218"},"PeriodicalIF":5.2,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142529771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Firefly forest: A swarm iteration-free swarm intelligence clustering algorithm 萤火虫森林无迭代群集智能聚类算法

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-18 DOI: 10.1016/j.jksuci.2024.102219

Shijie Zeng , Yuefei Wang , Yukun Wen , Xi Yu , Binxiong Li , Zixu Wang

The Firefly Forest algorithm is a novel bio-inspired clustering method designed to address key challenges in traditional clustering techniques, such as the need to set a fixed number of neighbors, predefine cluster numbers, and rely on computationally intensive swarm iterative processes. The algorithm begins by using an adaptive neighbor estimation, refined to filter outliers, to determine the brightness of each firefly. This brightness guides the formation of firefly trees, which are then merged into cohesive firefly forests, completing the clustering process. This approach allows the algorithm to dynamically capture both local and global patterns, eliminate the need for predefined cluster numbers, and operate with low computational complexity. Experiments involving 14 established clustering algorithms across 19 diverse datasets, using 8 evaluative metrics, demonstrate the Firefly Forest algorithm’s superior accuracy and robustness. These results highlight its potential as a powerful tool for real-world clustering applications. Our code is available at: https://github.com/firesaku/FireflyForest.

萤火虫森林算法是一种新颖的生物启发聚类方法，旨在解决传统聚类技术面临的主要挑战，如需要设置固定的邻居数量、预先确定聚类数量，以及依赖计算密集型的蜂群迭代过程。该算法首先使用自适应邻居估计，并对其进行改进以过滤异常值，从而确定每个萤火虫的亮度。这种亮度会引导萤火虫树的形成，然后将其合并成有凝聚力的萤火虫森林，完成聚类过程。这种方法允许算法动态捕捉局部和全局模式，无需预定义的聚类数量，并且计算复杂度低。在 19 个不同的数据集上使用 14 种成熟的聚类算法，并使用 8 个评估指标进行实验，结果表明萤火虫森林算法具有卓越的准确性和鲁棒性。这些结果凸显了萤火虫森林算法作为现实世界聚类应用的强大工具的潜力。我们的代码可在以下网址获取：https://github.com/firesaku/FireflyForest。

{"title":"Firefly forest: A swarm iteration-free swarm intelligence clustering algorithm","authors":"Shijie Zeng , Yuefei Wang , Yukun Wen , Xi Yu , Binxiong Li , Zixu Wang","doi":"10.1016/j.jksuci.2024.102219","DOIUrl":"10.1016/j.jksuci.2024.102219","url":null,"abstract":"<div><div>The Firefly Forest algorithm is a novel bio-inspired clustering method designed to address key challenges in traditional clustering techniques, such as the need to set a fixed number of neighbors, predefine cluster numbers, and rely on computationally intensive swarm iterative processes. The algorithm begins by using an adaptive neighbor estimation, refined to filter outliers, to determine the brightness of each firefly. This brightness guides the formation of firefly trees, which are then merged into cohesive firefly forests, completing the clustering process. This approach allows the algorithm to dynamically capture both local and global patterns, eliminate the need for predefined cluster numbers, and operate with low computational complexity. Experiments involving 14 established clustering algorithms across 19 diverse datasets, using 8 evaluative metrics, demonstrate the Firefly Forest algorithm’s superior accuracy and robustness. These results highlight its potential as a powerful tool for real-world clustering applications. Our code is available at: https://github.com/firesaku/FireflyForest.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102219"},"PeriodicalIF":5.2,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142529769","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IMOABC: An efficient multi-objective filter–wrapper hybrid approach for high-dimensional feature selection IMOABC：用于高维特征选择的高效多目标滤波器-包装器混合方法

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-09 DOI: 10.1016/j.jksuci.2024.102205

Jiahao Li , Tao Luo, Baitao Zhang, Min Chen, Jie Zhou

With the development of data science, the challenge of high-dimensional data has become increasingly prevalent. High-dimensional data contains a significant amount of redundant information, which can adversely affect the performance and effectiveness of machine learning algorithms. Therefore, it is necessary to select the most relevant features from the raw data and perform feature selection on high-dimensional data. In this paper, a novel filter–wrapper feature selection method based on an improved multi-objective artificial bee colony algorithm (IMOABC) is proposed to address the feature selection problem in high-dimensional data. This method simultaneously considers three objectives: feature error rate, feature subset ratio, and distance, effectively improving the efficiency of obtaining the optimal feature subset on high-dimensional data. Additionally, a novel Fisher Score-based initialization strategy is introduced, significantly enhancing the quality of solutions. Furthermore, a new dynamic adaptive strategy is designed, effectively improving the algorithm’s convergence speed and enhancing its global search capability. Comparative experimental results on microarray cancer datasets demonstrate that the proposed method significantly improves classification accuracy and effectively reduces the size of the feature subset when compared to various traditional and state-of-the-art multi-objective feature selection algorithms. IMOABC improves the classification accuracy by 2.27% on average compared to various multi-objective feature selection methods, while reducing the number of selected features by 88.76% on average. Meanwhile, IMOABC shows an average improvement of 4.24% in classification accuracy across all datasets, with an average reduction of 76.73% in the number of selected features compared to various traditional methods.

随着数据科学的发展，高维数据的挑战变得越来越普遍。高维数据包含大量冗余信息，会对机器学习算法的性能和效果产生不利影响。因此，有必要从原始数据中选择最相关的特征，并对高维数据进行特征选择。本文提出了一种基于改进的多目标人工蜂群算法（IMOABC）的新型滤波包特征选择方法，以解决高维数据中的特征选择问题。该方法同时考虑了特征误差率、特征子集比和距离三个目标，有效提高了在高维数据中获得最佳特征子集的效率。此外，该方法还引入了一种基于 Fisher Score 的新型初始化策略，大大提高了解决方案的质量。此外，还设计了一种新的动态自适应策略，有效提高了算法的收敛速度，增强了全局搜索能力。微阵列癌症数据集的对比实验结果表明，与各种传统和最先进的多目标特征选择算法相比，IMOABC 能显著提高分类准确率，并有效减少特征子集的大小。与各种多目标特征选择方法相比，IMOABC 的分类准确率平均提高了 2.27%，而所选特征的数量平均减少了 88.76%。同时，与各种传统方法相比，IMOABC 在所有数据集上的分类准确率平均提高了 4.24%，所选特征的数量平均减少了 76.73%。

{"title":"IMOABC: An efficient multi-objective filter–wrapper hybrid approach for high-dimensional feature selection","authors":"Jiahao Li , Tao Luo, Baitao Zhang, Min Chen, Jie Zhou","doi":"10.1016/j.jksuci.2024.102205","DOIUrl":"10.1016/j.jksuci.2024.102205","url":null,"abstract":"<div><div>With the development of data science, the challenge of high-dimensional data has become increasingly prevalent. High-dimensional data contains a significant amount of redundant information, which can adversely affect the performance and effectiveness of machine learning algorithms. Therefore, it is necessary to select the most relevant features from the raw data and perform feature selection on high-dimensional data. In this paper, a novel filter–wrapper feature selection method based on an improved multi-objective artificial bee colony algorithm (IMOABC) is proposed to address the feature selection problem in high-dimensional data. This method simultaneously considers three objectives: feature error rate, feature subset ratio, and distance, effectively improving the efficiency of obtaining the optimal feature subset on high-dimensional data. Additionally, a novel Fisher Score-based initialization strategy is introduced, significantly enhancing the quality of solutions. Furthermore, a new dynamic adaptive strategy is designed, effectively improving the algorithm’s convergence speed and enhancing its global search capability. Comparative experimental results on microarray cancer datasets demonstrate that the proposed method significantly improves classification accuracy and effectively reduces the size of the feature subset when compared to various traditional and state-of-the-art multi-objective feature selection algorithms. IMOABC improves the classification accuracy by 2.27% on average compared to various multi-objective feature selection methods, while reducing the number of selected features by 88.76% on average. Meanwhile, IMOABC shows an average improvement of 4.24% in classification accuracy across all datasets, with an average reduction of 76.73% in the number of selected features compared to various traditional methods.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102205"},"PeriodicalIF":5.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142432506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Advanced security measures in coupled phase-shift STAR-RIS networks: A DRL approach 耦合相移 STAR-RIS 网络中的高级安全措施：DRL 方法

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-09 DOI: 10.1016/j.jksuci.2024.102215

Abdul Wahid , Syed Zain Ul Abideen , Manzoor Ahmed , Wali Ullah Khan , Muhammad Sheraz , Teong Chee Chuah , Ying Loong Lee

The rapid development of next-generation wireless networks has intensified the need for robust security measures, particularly in environments susceptible to eavesdropping. Simultaneous transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) have emerged as a transformative technology, offering full-space coverage by manipulating electromagnetic wave propagation. However, the inherent flexibility of STAR-RIS introduces new vulnerabilities, making secure communication a significant challenge. To overcome these challenges, we propose a deep reinforcement learning (DRL) based secure and efficient beamforming optimization strategy, leveraging the deep deterministic policy gradient (DDPG) algorithm. By framing the problem as a Markov decision process (MDP), our approach enables the DDPG algorithm to learn optimal strategies for beamforming and transmission and reflection coefficients (TARCs) configurations. This method is specifically designed to optimize phase-shift coefficients within the STAR-RIS environment, effectively managing the coupled phase shifts and complex interactions that are critical for enhancing physical layer security (PLS). Through extensive simulations, we demonstrate that our DRL-based strategy not only outperforms traditional optimization techniques but also achieves real-time adaptive optimization, significantly improving both confidentiality and network efficiency. This research addresses key gaps in secure wireless communication and sets a new standard for future advancements in intelligent, adaptable network technologies.

下一代无线网络的快速发展加剧了对稳健安全措施的需求，尤其是在易被窃听的环境中。同时发射和反射可重构智能表面（STAR-RIS）作为一种变革性技术应运而生，通过操纵电磁波传播提供全空间覆盖。然而，STAR-RIS 固有的灵活性带来了新的漏洞，使安全通信成为一项重大挑战。为了克服这些挑战，我们利用深度确定性策略梯度（DDPG）算法，提出了一种基于深度强化学习（DRL）的安全高效波束成形优化策略。通过将问题框架化为马尔可夫决策过程（MDP），我们的方法使 DDPG 算法能够学习波束成形和传输与反射系数（TARC）配置的最佳策略。这种方法专为在 STAR-RIS 环境中优化相移系数而设计，可有效管理耦合相移和复杂的相互作用，这对增强物理层安全性（PLS）至关重要。通过大量仿真，我们证明了基于 DRL 的策略不仅优于传统优化技术，还能实现实时自适应优化，从而显著提高保密性和网络效率。这项研究填补了安全无线通信领域的关键空白，为未来智能、自适应网络技术的发展树立了新标准。

{"title":"Advanced security measures in coupled phase-shift STAR-RIS networks: A DRL approach","authors":"Abdul Wahid , Syed Zain Ul Abideen , Manzoor Ahmed , Wali Ullah Khan , Muhammad Sheraz , Teong Chee Chuah , Ying Loong Lee","doi":"10.1016/j.jksuci.2024.102215","DOIUrl":"10.1016/j.jksuci.2024.102215","url":null,"abstract":"<div><div>The rapid development of next-generation wireless networks has intensified the need for robust security measures, particularly in environments susceptible to eavesdropping. Simultaneous transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) have emerged as a transformative technology, offering full-space coverage by manipulating electromagnetic wave propagation. However, the inherent flexibility of STAR-RIS introduces new vulnerabilities, making secure communication a significant challenge. To overcome these challenges, we propose a deep reinforcement learning (DRL) based secure and efficient beamforming optimization strategy, leveraging the deep deterministic policy gradient (DDPG) algorithm. By framing the problem as a Markov decision process (MDP), our approach enables the DDPG algorithm to learn optimal strategies for beamforming and transmission and reflection coefficients (TARCs) configurations. This method is specifically designed to optimize phase-shift coefficients within the STAR-RIS environment, effectively managing the coupled phase shifts and complex interactions that are critical for enhancing physical layer security (PLS). Through extensive simulations, we demonstrate that our DRL-based strategy not only outperforms traditional optimization techniques but also achieves real-time adaptive optimization, significantly improving both confidentiality and network efficiency. This research addresses key gaps in secure wireless communication and sets a new standard for future advancements in intelligent, adaptable network technologies.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102215"},"PeriodicalIF":5.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142432505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Endoscopic video aided identification method for gastric area 内窥镜视频辅助胃区识别方法

IF 5.2 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Journal of King Saud University-Computer and Information Sciences

Pub Date : 2024-10-05 DOI: 10.1016/j.jksuci.2024.102208

Xiangwei Zheng, Dejian Su, Xuanchi Chen, Mingzhe Zhang

Probe-based confocal laser endomicroscopy (pCLE) is a significant diagnostic instrument and is frequently utilized to diagnose the severity of gastric intestinal metaplasia (GIM). The physicians must comprehensively analyze video clips recorded with pCLE from the gastric antrum, gastric body, and gastric angle area to determine the patient’s condition. However, due to the limitations of the pCLE’s microscopic imaging structure, the gastric areas detected cannot be identified and recorded in real time, which may poses a risk of missing potential areas of disease occurrence and is not conducive to the subsequent precise treatment of the lesion area. Therefore, this paper proposes an endoscopic video aided identification method for identifying gastric areas (EVIGA), which are utilized for determining the detected areas of pCLE in real-time. Firstly, the start time of the diagnosis clip is determined by real-time detecting the working states of pCLE. Then, the endoscopic video clip is truncated according to the correspondence between pCLE and endoscopic video in the time sequence for detecting gastric areas. In order to accurately identify pCLE detected gastric areas, a probe-based confocal laser endomicroscopy diagnosis area identification model (pCLEDAM) is constructed with an hourglass convolution designed for single-frame feature extraction and a temporal feature-sensitive extraction structure for spatial feature extraction. The extracted feature maps are unfolded and fed into the fully connected layer to classify the detected areas. To validate the proposed method, 67 clinical confocal laser endomicroscopy diagnosis cases are collected from a tertiary care hospital, and 500 video clips are finally reserved after audited for dataset construction. Experiments show that the accuracy of area identification on the test dataset achieves 96.0% and is much higher than other related algorithms, achieving the accurate identification of pCLE detected areas.

探针共焦激光内窥镜（pCLE）是一种重要的诊断仪器，经常被用来诊断胃肠变性（GIM）的严重程度。医生必须全面分析 pCLE 从胃窦、胃体和胃角区域记录的视频片段，以确定患者的病情。然而，由于pCLE显微成像结构的局限性，所检测到的胃部区域无法被实时识别和记录，有可能遗漏潜在的疾病发生区域，不利于后续对病变区域的精确治疗。因此，本文提出了一种内镜视频辅助胃区识别方法（EVIGA），用于实时确定检测到的胃癌病变区域。首先，通过实时检测 pCLE 的工作状态来确定诊断片段的开始时间。然后，根据 pCLE 和内窥镜视频在时间序列上的对应关系截断内窥镜视频片段，以检测胃部区域。为了准确识别 pCLE 检测到的胃区，构建了一个基于探针的共聚焦激光内窥镜诊断区域识别模型（pCLEDAM），其沙漏卷积设计用于单帧特征提取，时间特征敏感提取结构用于空间特征提取。提取的特征图被展开并输入全连接层，对检测到的区域进行分类。为了验证所提出的方法，从一家三甲医院收集了 67 个临床共焦激光内窥镜诊断病例，经审核后最终保留了 500 个视频片段用于数据集构建。实验表明，测试数据集的区域识别准确率达到 96.0%，远高于其他相关算法，实现了对 pCLE 检测区域的准确识别。

{"title":"Endoscopic video aided identification method for gastric area","authors":"Xiangwei Zheng, Dejian Su, Xuanchi Chen, Mingzhe Zhang","doi":"10.1016/j.jksuci.2024.102208","DOIUrl":"10.1016/j.jksuci.2024.102208","url":null,"abstract":"<div><div>Probe-based confocal laser endomicroscopy (pCLE) is a significant diagnostic instrument and is frequently utilized to diagnose the severity of gastric intestinal metaplasia (GIM). The physicians must comprehensively analyze video clips recorded with pCLE from the gastric antrum, gastric body, and gastric angle area to determine the patient’s condition. However, due to the limitations of the pCLE’s microscopic imaging structure, the gastric areas detected cannot be identified and recorded in real time, which may poses a risk of missing potential areas of disease occurrence and is not conducive to the subsequent precise treatment of the lesion area. Therefore, this paper proposes an endoscopic video aided identification method for identifying gastric areas (EVIGA), which are utilized for determining the detected areas of pCLE in real-time. Firstly, the start time of the diagnosis clip is determined by real-time detecting the working states of pCLE. Then, the endoscopic video clip is truncated according to the correspondence between pCLE and endoscopic video in the time sequence for detecting gastric areas. In order to accurately identify pCLE detected gastric areas, a probe-based confocal laser endomicroscopy diagnosis area identification model (pCLEDAM) is constructed with an hourglass convolution designed for single-frame feature extraction and a temporal feature-sensitive extraction structure for spatial feature extraction. The extracted feature maps are unfolded and fed into the fully connected layer to classify the detected areas. To validate the proposed method, 67 clinical confocal laser endomicroscopy diagnosis cases are collected from a tertiary care hospital, and 500 video clips are finally reserved after audited for dataset construction. Experiments show that the accuracy of area identification on the test dataset achieves 96.0% and is much higher than other related algorithms, achieving the accurate identification of pCLE detected areas.</div></div>","PeriodicalId":48547,"journal":{"name":"Journal of King Saud University-Computer and Information Sciences","volume":"36 9","pages":"Article 102208"},"PeriodicalIF":5.2,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142424361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0