Array最新文献_第2页

Agentic artificial intelligence is the future of cancer detection and diagnosis 人工智能是癌症检测和诊断的未来

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-31 DOI: 10.1016/j.array.2025.100676

Sayedur Rahman , Md. Tanzib Hosain , Nafiz Fahad , Md. Kishor Morol , Md. Jakir Hossen

Agentic artificial intelligence systems, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs), are a big change in oncology because they can find and diagnose cancer in ways that have never been done before. In accordance with PRISMA 2020 criteria, we conducted a systematic search across nine databases from January 2023 to September 2025, reviewing 3986 records and incorporating 123 papers that assessed agentic AI in cancer detection and diagnosis. Research demonstrated swift expansion (91.9% published in 2024-2025) across various cancer kinds, with breast (22.0%) and lung cancer (13.8%) being the most extensively examined. GPT-4 versions showed performance similar to that of human experts: they found errors better than pathologists (89.5% vs. 88.5%), classified skin lesions as well as dermatologists (84.8% vs. 84.6%), and staged ovarian cancer with 97% accuracy compared to 88% by radiologists. Zero-shot LLMs consistently surpassed conventional supervised models. But there were big problems, like factual errors in 15%–41% of instances, algorithmic bias, and low agreement with tumor boards (50%–70%). Agentic AI has a lot of promise for finding cancer, especially in organized tasks. However, the research so far suggests that it should be used as an aid rather than an independent system. Concerns about reliability and bias in algorithms are two of the most important impediments. Future priorities encompass Retrieval-Augmented Generation(RAG) systems, domain-specific models, and forthcoming trials to ascertain clinical value.

人工智能系统，特别是大型语言模型（llm）和视觉语言模型（vlm），是肿瘤学领域的一个重大变革，因为它们可以以前所未有的方式发现和诊断癌症。根据PRISMA 2020标准，我们在2023年1月至2025年9月期间对9个数据库进行了系统检索，审查了3986条记录，并纳入了123篇评估人工智能在癌症检测和诊断中的应用的论文。研究表明，在各种癌症类型中，乳腺癌（22.0%）和肺癌（13.8%）的研究范围迅速扩大（2024-2025年发表了91.9%）。GPT-4版本的表现与人类专家相似：它们比病理学家更容易发现错误（89.5%对88.5%），对皮肤病变的分类和皮肤科医生一样好（84.8%对84.6%），对卵巢癌的分期准确率为97%，而放射科医生的准确率为88%。零射击llm始终优于传统的监督模型。但也存在一些大问题，比如15%-41%的案例存在事实错误、算法偏差以及与肿瘤委员会的一致性较低（50%-70%）。人工智能在发现癌症方面有很大的前景，尤其是在有组织的任务中。然而，迄今为止的研究表明，它应该被用作一种辅助手段，而不是一个独立的系统。对可靠性和算法偏差的担忧是两个最重要的障碍。未来的优先事项包括检索增强生成（RAG）系统、领域特定模型和即将进行的试验，以确定临床价值。

{"title":"Agentic artificial intelligence is the future of cancer detection and diagnosis","authors":"Sayedur Rahman , Md. Tanzib Hosain , Nafiz Fahad , Md. Kishor Morol , Md. Jakir Hossen","doi":"10.1016/j.array.2025.100676","DOIUrl":"10.1016/j.array.2025.100676","url":null,"abstract":"<div><div>Agentic artificial intelligence systems, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs), are a big change in oncology because they can find and diagnose cancer in ways that have never been done before. In accordance with PRISMA 2020 criteria, we conducted a systematic search across nine databases from January 2023 to September 2025, reviewing 3986 records and incorporating 123 papers that assessed agentic AI in cancer detection and diagnosis. Research demonstrated swift expansion (91.9% published in 2024-2025) across various cancer kinds, with breast (22.0%) and lung cancer (13.8%) being the most extensively examined. GPT-4 versions showed performance similar to that of human experts: they found errors better than pathologists (89.5% vs. 88.5%), classified skin lesions as well as dermatologists (84.8% vs. 84.6%), and staged ovarian cancer with 97% accuracy compared to 88% by radiologists. Zero-shot LLMs consistently surpassed conventional supervised models. But there were big problems, like factual errors in 15%–41% of instances, algorithmic bias, and low agreement with tumor boards (50%–70%). Agentic AI has a lot of promise for finding cancer, especially in organized tasks. However, the research so far suggests that it should be used as an aid rather than an independent system. Concerns about reliability and bias in algorithms are two of the most important impediments. Future priorities encompass Retrieval-Augmented Generation(RAG) systems, domain-specific models, and forthcoming trials to ascertain clinical value.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100676"},"PeriodicalIF":4.5,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921324","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

Smart driving with AI: A review of CNN approaches to drowsiness detection 人工智能智能驾驶：CNN睡意检测方法综述

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-31 DOI: 10.1016/j.array.2025.100675

Riadul Islam Rabbi , Poh Ping Em , Md. Jakir Hossen

Drowsy driving is widespread and a significant cause of traffic accidents, and thus poses a serious threat to life and property around the globe. Therefore, real-time driver drowsiness detection has emerged as a primary study area, particularly due to the current advancements that incorporate artificial intelligence (AI) into automobiles. Convolutional Neural Networks (CNNs) have recently been very effective in handling image data and feature extraction for detecting drowsiness based on facial and eye movement patterns. This review paper focuses on the different CNN architectures and models that exist in the field of driver drowsiness detection and their strengths and limitations. Models like VGGNet, ResNet, and Inception V3 that are used in CNN are elaborated using pseudocode for an easy understanding of how they can be implemented practically. This paper also examines new trends in lightweight CNNs for edge computing as a solution to demands for real-time analytics in constrained environments such as vehicles. Moreover, important issues like data bias, model overfitting, and computational constraints are discussed. Additionally, future perspectives are provided to address these challenges, such as the integration of hybrid models and fusion of multimodal data. This review aims to provide a comprehensive understanding of CNN-based drowsiness detection and assist in developing safe and reliable automotive applications.

疲劳驾驶普遍存在，是造成交通事故的重要原因，在全球范围内对生命财产构成严重威胁。因此，实时驾驶员困倦检测已经成为一个主要的研究领域，特别是由于目前将人工智能（AI）融入汽车的进步。卷积神经网络（cnn）最近在处理图像数据和基于面部和眼球运动模式检测睡意的特征提取方面非常有效。本文主要介绍了目前存在于驾驶员困倦检测领域的不同CNN架构和模型，以及它们的优势和局限性。在CNN中使用的VGGNet、ResNet和Inception V3等模型都是使用伪代码进行阐述的，以便于理解如何在实际中实现它们。本文还研究了用于边缘计算的轻量级cnn的新趋势，作为解决车辆等受限环境中实时分析需求的解决方案。此外，还讨论了数据偏差、模型过拟合和计算约束等重要问题。此外，还提供了解决这些挑战的未来展望，例如混合模型的集成和多模态数据的融合。本文旨在全面了解基于cnn的困倦检测，并协助开发安全可靠的汽车应用。

{"title":"Smart driving with AI: A review of CNN approaches to drowsiness detection","authors":"Riadul Islam Rabbi , Poh Ping Em , Md. Jakir Hossen","doi":"10.1016/j.array.2025.100675","DOIUrl":"10.1016/j.array.2025.100675","url":null,"abstract":"<div><div>Drowsy driving is widespread and a significant cause of traffic accidents, and thus poses a serious threat to life and property around the globe. Therefore, real-time driver drowsiness detection has emerged as a primary study area, particularly due to the current advancements that incorporate artificial intelligence (AI) into automobiles. Convolutional Neural Networks (CNNs) have recently been very effective in handling image data and feature extraction for detecting drowsiness based on facial and eye movement patterns. This review paper focuses on the different CNN architectures and models that exist in the field of driver drowsiness detection and their strengths and limitations. Models like VGGNet, ResNet, and Inception V3 that are used in CNN are elaborated using pseudocode for an easy understanding of how they can be implemented practically. This paper also examines new trends in lightweight CNNs for edge computing as a solution to demands for real-time analytics in constrained environments such as vehicles. Moreover, important issues like data bias, model overfitting, and computational constraints are discussed. Additionally, future perspectives are provided to address these challenges, such as the integration of hybrid models and fusion of multimodal data. This review aims to provide a comprehensive understanding of CNN-based drowsiness detection and assist in developing safe and reliable automotive applications.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100675"},"PeriodicalIF":4.5,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921318","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

Trans-ADENet: Transformer-based Attention-guided Deep Ensemble Network for high-dimensional data classification Trans-ADENet：用于高维数据分类的基于变压器的注意力引导深度集成网络

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-30 DOI: 10.1016/j.array.2025.100659

Venkaiah Chowdary Bhimineni, Rajiv Senapati

High-dimensional (HD) biomedical data, such as gene expression profiles and ECG signals, pose significant challenges for machine learning (ML) due to limited sample size, feature redundancy, and noisy distributions. Conventional models tend to overfit, whereas boosting and ensemble approaches struggle with irrelevant features. Deep autoencoders (DAE) reduce nonlinear dimensionality but miss complex dependencies, whereas transformers require large datasets to model long-range relationships through self-attention mechanisms. We propose a Transformer-based Attention-guided Deep Ensemble Network (Trans-ADENet) that integrates dimensionality reduction, attention-driven feature learning, and meta-level ensemble fusion in an end-to-end framework. A deep autoencoder compresses HD inputs into compact latent representations, refined by a Transformer Encoder with multi-head self-attention. Refined features are fed to diverse base classifiers (CatBoost, Support Vector Machine (SVM), TabNet, and Generalized Multi-Layer Perceptron (GMLP)), and their outputs are fused by a meta-MLP, which learns adaptive weights to yield robust predictions. Experiments on breast, leukemia, INCART2 and Thyroid-RNA datasets achieved 96.3%, 94.1%, 92.7% and 94.6% accuracy, surpassing state-of-the-art models in terms of accuracy, F1, precision, recall, and AUC. By combining representation learning, attention, and adaptive fusion, Trans-ADENet delivers accurate, interpretable classification for biomedical tasks.

高维（HD）生物医学数据，如基因表达谱和心电信号，由于样本量有限、特征冗余和噪声分布，对机器学习（ML）构成了重大挑战。传统模型倾向于过度拟合，而增强和集成方法则与不相关的特征作斗争。深度自编码器（DAE）减少了非线性维度，但忽略了复杂的依赖关系，而变压器需要大型数据集来通过自关注机制建模长期关系。我们提出了一种基于transformer的注意力引导深度集成网络（Trans-ADENet），该网络在端到端框架中集成了降维、注意力驱动特征学习和元级集成融合。深度自编码器将HD输入压缩成紧凑的潜在表示，由具有多头自关注的变压器编码器进行细化。将精炼的特征输入到不同的基本分类器（CatBoost、支持向量机（SVM）、TabNet和广义多层感知器（GMLP））中，它们的输出由元mlp融合，该元mlp学习自适应权重以产生稳健的预测。在乳腺癌、白血病、INCART2和Thyroid-RNA数据集上的实验，准确率分别达到96.3%、94.1%、92.7%和94.6%，在准确率、F1、精密度、召回率和AUC方面都超过了目前最先进的模型。通过结合表征学习、注意和自适应融合，Trans-ADENet为生物医学任务提供了准确的、可解释的分类。

{"title":"Trans-ADENet: Transformer-based Attention-guided Deep Ensemble Network for high-dimensional data classification","authors":"Venkaiah Chowdary Bhimineni, Rajiv Senapati","doi":"10.1016/j.array.2025.100659","DOIUrl":"10.1016/j.array.2025.100659","url":null,"abstract":"<div><div>High-dimensional (HD) biomedical data, such as gene expression profiles and ECG signals, pose significant challenges for machine learning (ML) due to limited sample size, feature redundancy, and noisy distributions. Conventional models tend to overfit, whereas boosting and ensemble approaches struggle with irrelevant features. Deep autoencoders (DAE) reduce nonlinear dimensionality but miss complex dependencies, whereas transformers require large datasets to model long-range relationships through self-attention mechanisms. We propose a Transformer-based Attention-guided Deep Ensemble Network (Trans-ADENet) that integrates dimensionality reduction, attention-driven feature learning, and meta-level ensemble fusion in an end-to-end framework. A deep autoencoder compresses HD inputs into compact latent representations, refined by a Transformer Encoder with multi-head self-attention. Refined features are fed to diverse base classifiers (CatBoost, Support Vector Machine (SVM), TabNet, and Generalized Multi-Layer Perceptron (GMLP)), and their outputs are fused by a meta-MLP, which learns adaptive weights to yield robust predictions. Experiments on breast, leukemia, INCART2 and Thyroid-RNA datasets achieved 96.3%, 94.1%, 92.7% and 94.6% accuracy, surpassing state-of-the-art models in terms of accuracy, F1, precision, recall, and AUC. By combining representation learning, attention, and adaptive fusion, Trans-ADENet delivers accurate, interpretable classification for biomedical tasks.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100659"},"PeriodicalIF":4.5,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921394","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

Federated Convolutional Neural Networks (F-CNNs) for privacy-preserving multi-class skin lesion classification 联邦卷积神经网络（f - cnn）用于保护隐私的多类皮肤病变分类

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-30 DOI: 10.1016/j.array.2025.100667

Khadija Shahzad , Anum Khashir , Hina Tufail , Abdul Ahad , Zahra Ali , Filipe Madeira , Ivan Miguel Pires

Skin lesions include a variety of abnormalities found on the skin. These may be benign (not cancerous) or malignant (cancerous). Every year, the number of cases of skin cancer increases globally, increasing the death rate. Medical data is scarce because people are reluctant to provide their health information due to privacy concerns. In this research, a decentralized machine learning approach, Federated learning, is the primary focus of the discipline to preserve patient data. Using this method, models are trained independently on several dispersed devices without sharing the data. To balance the data and enrich the dataset, the Synthetic Minority Over-sampling technique with Edited Nearest Neighbors (SMOTEENN) is used in this study. The HAM10000 dataset was benchmarked using a Convolutional Neural Network (CNN). Seven classes of HAM10000 include vascular skin lesions, benign keratosis, actinic keratosis, melanoma, dermatofibroma, and melanocytic nevi. A centralized method yields an accuracy of 99.39%, and f1-score, precision, and recall of 99.00%. A simulated Federated learning with three clients, ten rounds, and thirty training epochs produced 93.00% precision, 92.00% recall, 92.00% f1-score, and 91.80% accuracy, respectively. At the same time, an increase to four clients and thirty training epochs produced an accuracy, recall, precision, and f1-score of 97.00% with ten rounds.

皮肤病变包括在皮肤上发现的各种异常。这些可能是良性的（非癌性的）或恶性的（癌性的）。每年，全球皮肤癌病例数都在增加，死亡率也在增加。医疗数据很少，因为人们出于隐私考虑不愿提供自己的健康信息。在这项研究中，分散的机器学习方法，联邦学习，是该学科保存患者数据的主要焦点。使用该方法，模型在多个分散的设备上独立训练，而不共享数据。为了平衡数据和丰富数据集，本研究使用了编辑近邻的合成少数派过采样技术（SMOTEENN）。HAM10000数据集使用卷积神经网络（CNN）进行基准测试。HAM10000分为7类：血管性皮肤病变、良性角化病、光化性角化病、黑色素瘤、皮肤纤维瘤、黑素细胞痣。集中式方法的准确率为99.39%，f1-score、精密度和召回率为99.00%。模拟的联邦学习有3个客户端、10轮和30个训练周期，分别产生了93.00%的准确率、92.00%的召回率、92.00%的f1得分和91.80%的准确率。与此同时，增加到4个客户和30个训练周期，在10轮训练中，准确率、召回率、精确度和f1得分达到97.00%。

{"title":"Federated Convolutional Neural Networks (F-CNNs) for privacy-preserving multi-class skin lesion classification","authors":"Khadija Shahzad , Anum Khashir , Hina Tufail , Abdul Ahad , Zahra Ali , Filipe Madeira , Ivan Miguel Pires","doi":"10.1016/j.array.2025.100667","DOIUrl":"10.1016/j.array.2025.100667","url":null,"abstract":"<div><div>Skin lesions include a variety of abnormalities found on the skin. These may be benign (not cancerous) or malignant (cancerous). Every year, the number of cases of skin cancer increases globally, increasing the death rate. Medical data is scarce because people are reluctant to provide their health information due to privacy concerns. In this research, a decentralized machine learning approach, Federated learning, is the primary focus of the discipline to preserve patient data. Using this method, models are trained independently on several dispersed devices without sharing the data. To balance the data and enrich the dataset, the Synthetic Minority Over-sampling technique with Edited Nearest Neighbors (SMOTEENN) is used in this study. The HAM10000 dataset was benchmarked using a Convolutional Neural Network (CNN). Seven classes of HAM10000 include vascular skin lesions, benign keratosis, actinic keratosis, melanoma, dermatofibroma, and melanocytic nevi. A centralized method yields an accuracy of 99.39%, and f1-score, precision, and recall of 99.00%. A simulated Federated learning with three clients, ten rounds, and thirty training epochs produced 93.00% precision, 92.00% recall, 92.00% f1-score, and 91.80% accuracy, respectively. At the same time, an increase to four clients and thirty training epochs produced an accuracy, recall, precision, and f1-score of 97.00% with ten rounds.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100667"},"PeriodicalIF":4.5,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973363","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

Weighted neural network for imbalanced information with undersampling 欠采样不平衡信息的加权神经网络

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-29 DOI: 10.1016/j.array.2025.100668

Simegnew Yihunie Alaba , Nahian A. Siddique , Emrah Benli , Dominik Enns , Yuichi Motai

Class imbalance presents a significant challenge in machine learning, particularly in applications where accurate detection of minority classes is essential. This study presents an innovative method for precisely and sensitively classifying imbalanced data using a feed-forward artificial neural network architecture. By addressing the limitations of conventional linear models, which often overlook minority classes, the proposed method combines selective undersampling, adaptive kernel optimization, and a weighted neural network structure. The training dataset is iteratively refined to prioritize minority classes, while a kernel function optimizer sharpens class boundaries through adaptive transformations. Experimental results demonstrate that the proposed model consistently outperforms standard techniques on diverse imbalanced datasets, achieving high accuracy (above 90 %), sensitivity, and G-mean scores (above 80 %). Computational efficiency analysis further strengthens the method’s practicality for real-world applications by balancing classification precision with processing efficiency.

类不平衡在机器学习中提出了重大挑战，特别是在精确检测少数类至关重要的应用中。本文提出了一种利用前馈人工神经网络结构对不平衡数据进行精确、灵敏分类的创新方法。通过解决传统线性模型经常忽略少数类的局限性，该方法结合了选择性欠采样、自适应核优化和加权神经网络结构。训练数据集迭代细化以优先考虑少数类，而核函数优化器通过自适应转换锐化类边界。实验结果表明，该模型在不同的不平衡数据集上始终优于标准技术，实现了较高的准确性（90%以上）、灵敏度和g均值得分（80%以上）。计算效率分析通过平衡分类精度和处理效率，进一步增强了该方法在实际应用中的实用性。

引用次数: 0

Gaze-adaptive neural pre-correction for mitigating spatially varying optical aberrations in near-eye displays 用于减轻近眼显示中空间变化光学像差的注视自适应神经预校正

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-27 DOI: 10.1016/j.array.2025.100654

Yi Jiang, Ye Bi, Yinng Li, Pengfei Li, Shengnan Qin, Zichao Shu, Chengrui Le

Near-eye display (NED) technology constitutes a fundamental component of head-mounted display (HMD) systems. The compact form factor required by HMDs imposes stringent constraints on optical design, often resulting in pronounced wavefront aberrations that significantly degrade visual fidelity. In addition, natural eye movements dynamically induce varying blur that further compromises image quality. To mitigate these challenges, a gaze-contingent neural network framework has been developed to compensate for aberrations within the foveal region. The network is trained in an end-to-end manner to minimize the discrepancy between the optically degraded system output and the corresponding ground truth image. A forward imaging model is employed, in which the network output is convolved with a spatially varying point spread function (PSF) to accurately simulate the degradation introduced by the optical system. To accommodate dynamic changes in gaze direction, a foveated attention-guided module is incorporated to adaptively modulate the pre-correction process, enabling localized compensation centered on the fovea. Additionally, an end-to-end trainable architecture has been designed to integrate gaze-informed blur priors. Both simulation and experimental validations confirm that the proposed method substantially reduces gaze-dependent aberrations and enhances retinal image clarity within the foveal region, while maintaining high computational efficiency. The presented framework offers a practical and scalable solution for improving visual performance in aberration-sensitive NED systems.

近眼显示（NED）技术是头戴式显示（HMD）系统的基本组成部分。hmd所要求的紧凑外形因素对光学设计施加了严格的限制，通常会导致明显的波前像差，从而显著降低视觉保真度。此外，自然的眼球运动动态地诱导不同的模糊，进一步损害图像质量。为了减轻这些挑战，研究人员开发了一种基于注视的神经网络框架来补偿中央凹区域内的像差。该网络以端到端方式进行训练，以最小化光学退化系统输出与相应的地面真值图像之间的差异。采用前向成像模型，将网络输出与空间变化点扩展函数（PSF）进行卷积，以精确模拟光学系统引入的退化。为了适应注视方向的动态变化，我们采用了一个注视点注意引导模块来自适应地调节预校正过程，从而实现以中央凹为中心的局部补偿。此外，端到端可训练的架构已被设计集成的视线通知模糊先验。仿真和实验验证表明，该方法在保持较高的计算效率的同时，大大降低了注视相关像差，提高了中央凹区域内视网膜图像的清晰度。所提出的框架为改善像差敏感NED系统的视觉性能提供了一个实用且可扩展的解决方案。

{"title":"Gaze-adaptive neural pre-correction for mitigating spatially varying optical aberrations in near-eye displays","authors":"Yi Jiang, Ye Bi, Yinng Li, Pengfei Li, Shengnan Qin, Zichao Shu, Chengrui Le","doi":"10.1016/j.array.2025.100654","DOIUrl":"10.1016/j.array.2025.100654","url":null,"abstract":"<div><div>Near-eye display (NED) technology constitutes a fundamental component of head-mounted display (HMD) systems. The compact form factor required by HMDs imposes stringent constraints on optical design, often resulting in pronounced wavefront aberrations that significantly degrade visual fidelity. In addition, natural eye movements dynamically induce varying blur that further compromises image quality. To mitigate these challenges, a gaze-contingent neural network framework has been developed to compensate for aberrations within the foveal region. The network is trained in an end-to-end manner to minimize the discrepancy between the optically degraded system output and the corresponding ground truth image. A forward imaging model is employed, in which the network output is convolved with a spatially varying point spread function (PSF) to accurately simulate the degradation introduced by the optical system. To accommodate dynamic changes in gaze direction, a foveated attention-guided module is incorporated to adaptively modulate the pre-correction process, enabling localized compensation centered on the fovea. Additionally, an end-to-end trainable architecture has been designed to integrate gaze-informed blur priors. Both simulation and experimental validations confirm that the proposed method substantially reduces gaze-dependent aberrations and enhances retinal image clarity within the foveal region, while maintaining high computational efficiency. The presented framework offers a practical and scalable solution for improving visual performance in aberration-sensitive NED systems.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100654"},"PeriodicalIF":4.5,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921260","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

Leveraging blockchain and LLMs for patient–clinical trial matching 利用区块链和llm进行患者-临床试验匹配

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-27 DOI: 10.1016/j.array.2025.100657

Diana Hawashin , Khaled Salah , Raja Jayaraman , Samer Ellahham , Ibrar Yaqoob

Efficiently matching patients to clinical trials is essential for advancing medical research and ensuring reliable outcomes. However, current matching methods face several challenges. These include data integrity issues from tampered records, privacy risks caused by weak anonymization, and manual processes that delay recruitment. In addition, centralized systems lack transparency, expose sensitive patient data to security vulnerabilities, and suffer from single points of failure that reduce resilience and trust. In this paper, we propose a blockchain and Large Language Models (LLMs)-driven solution for secure, trustworthy, traceable, decentralized, and transparent patient–clinical trial matching. Blockchain ensures data integrity, security, and transparency by eliminating single points of failure and enabling tamper-proof records. LLMs enhance patient–trial matching by automating the interpretation of complex eligibility criteria, improving accuracy, and significantly reducing the time required for manual review. Our approach uses Ethereum-based smart contracts to automate workflows such as trial registration, eligibility assessment, and consent tracking. We fine-tune GPT-4, T5, and Gemini on synthetic data derived from real clinical trial records and employ majority voting to ensure consistent and unbiased eligibility decisions. A prototype Gradio interface was developed as a minimum viable product (MVP) to demonstrate seamless interaction between LLMs and smart contracts. Performance evaluation based on accuracy (0.800), precision (0.733), recall (1.000), and F1-score (0.846) demonstrates reliable eligibility prediction. Cost analysis confirms affordability, and security evaluation verifies resilience against known threats. Comparison with existing solutions highlights the framework’s advantages in transparency, trust, and automation. The smart contract code is publicly available on GitHub.

有效地将患者与临床试验相匹配，对于推进医学研究和确保可靠的结果至关重要。然而，目前的匹配方法面临着一些挑战。这些问题包括篡改记录造成的数据完整性问题、弱匿名化造成的隐私风险，以及延迟招聘的手动流程。此外，集中式系统缺乏透明度，将敏感的患者数据暴露给安全漏洞，并且存在单点故障，从而降低了弹性和信任。在本文中，我们提出了一个区块链和大型语言模型（llm）驱动的解决方案，用于安全、可信、可追溯、分散和透明的患者-临床试验匹配。区块链通过消除单点故障和启用防篡改记录来确保数据的完整性、安全性和透明性。llm通过自动化解释复杂的资格标准，提高准确性，并显着减少人工审查所需的时间，从而增强了患者-试验匹配。我们的方法使用基于以太坊的智能合约来自动化工作流程，如试验注册，资格评估和同意跟踪。我们根据来自真实临床试验记录的合成数据对GPT-4、T5和Gemini进行微调，并采用多数投票来确保一致和公正的资格决定。作为最小可行产品（MVP），开发了一个原型gradient接口，以演示llm和智能合约之间的无缝交互。基于正确率（0.800）、精密度（0.733）、召回率（1.000）和f1分数（0.846）的性能评价表明合格性预测是可靠的。成本分析确认可负担性，安全评估验证针对已知威胁的弹性。与现有解决方案的比较突出了该框架在透明度、信任和自动化方面的优势。智能合约代码在GitHub上公开可用。

{"title":"Leveraging blockchain and LLMs for patient–clinical trial matching","authors":"Diana Hawashin , Khaled Salah , Raja Jayaraman , Samer Ellahham , Ibrar Yaqoob","doi":"10.1016/j.array.2025.100657","DOIUrl":"10.1016/j.array.2025.100657","url":null,"abstract":"<div><div>Efficiently matching patients to clinical trials is essential for advancing medical research and ensuring reliable outcomes. However, current matching methods face several challenges. These include data integrity issues from tampered records, privacy risks caused by weak anonymization, and manual processes that delay recruitment. In addition, centralized systems lack transparency, expose sensitive patient data to security vulnerabilities, and suffer from single points of failure that reduce resilience and trust. In this paper, we propose a blockchain and Large Language Models (LLMs)-driven solution for secure, trustworthy, traceable, decentralized, and transparent patient–clinical trial matching. Blockchain ensures data integrity, security, and transparency by eliminating single points of failure and enabling tamper-proof records. LLMs enhance patient–trial matching by automating the interpretation of complex eligibility criteria, improving accuracy, and significantly reducing the time required for manual review. Our approach uses Ethereum-based smart contracts to automate workflows such as trial registration, eligibility assessment, and consent tracking. We fine-tune GPT-4, T5, and Gemini on synthetic data derived from real clinical trial records and employ majority voting to ensure consistent and unbiased eligibility decisions. A prototype Gradio interface was developed as a minimum viable product (MVP) to demonstrate seamless interaction between LLMs and smart contracts. Performance evaluation based on accuracy (0.800), precision (0.733), recall (1.000), and F1-score (0.846) demonstrates reliable eligibility prediction. Cost analysis confirms affordability, and security evaluation verifies resilience against known threats. Comparison with existing solutions highlights the framework’s advantages in transparency, trust, and automation. The smart contract code is publicly available on GitHub.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100657"},"PeriodicalIF":4.5,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921257","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

AgnoSVD: Dynamic resource allocation for serverless workloads using collaborative filtering AgnoSVD：使用协作过滤为无服务器工作负载进行动态资源分配

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-27 DOI: 10.1016/j.array.2025.100662

Md. Shariar Kabir, Muhammad Abdullah Adnan

In serverless computing, determining the optimal resource configurations for workloads poses significant challenges, particularly due to the cloud provider’s limited visibility into workload specifics. This complexity is amplified when dealing with diverse workloads that vary in their characteristics. In this paper, we present AgnoSVD, an approach for predicting the optimum resource configuration for an incoming workload using Singular Value Decomposition (SVD). The proposed model uses collaborative filtering to extract the latent factors of the workloads and resource profiles. Therefore, the model remains agnostic to the specific details of the functions and the resource configurations. We tested our approach on well-known serverless systems like AWS lambda and Apache OpenWhisk and evaluated the system using 99 functional workloads. These workloads encompass both individual functions and chains of functions, addressing a range of computational and learning problems. To validate the system’s ability to adapt to changes, we also evaluated our system using functions with different input parameter sizes. Our evaluation shows that the model reaches convergence within 2 feedback iterations and results in a 32.41% decrease in average cost and a 5.18% average speedup, outperforming other state-of-the-art approaches.

在无服务器计算中，确定工作负载的最佳资源配置带来了重大挑战，特别是由于云提供商对工作负载细节的可见性有限。在处理特征各异的各种工作负载时，这种复杂性会被放大。在本文中，我们提出了AgnoSVD，这是一种使用奇异值分解（SVD）预测传入工作负载的最佳资源配置的方法。该模型采用协同过滤的方法提取工作负载和资源配置文件的潜在因素。因此，模型对功能和资源配置的具体细节保持不可知。我们在知名的无服务器系统（如AWS lambda和Apache OpenWhisk）上测试了我们的方法，并使用99个功能工作负载对系统进行了评估。这些工作负载包括单个功能和功能链，解决一系列计算和学习问题。为了验证系统适应变化的能力，我们还使用具有不同输入参数大小的函数来评估我们的系统。我们的评估表明，该模型在2次反馈迭代内达到收敛，平均成本降低32.41%，平均加速提高5.18%，优于其他最先进的方法。

{"title":"AgnoSVD: Dynamic resource allocation for serverless workloads using collaborative filtering","authors":"Md. Shariar Kabir, Muhammad Abdullah Adnan","doi":"10.1016/j.array.2025.100662","DOIUrl":"10.1016/j.array.2025.100662","url":null,"abstract":"<div><div>In serverless computing, determining the optimal resource configurations for workloads poses significant challenges, particularly due to the cloud provider’s limited visibility into workload specifics. This complexity is amplified when dealing with diverse workloads that vary in their characteristics. In this paper, we present AgnoSVD, an approach for predicting the optimum resource configuration for an incoming workload using Singular Value Decomposition (SVD). The proposed model uses collaborative filtering to extract the latent factors of the workloads and resource profiles. Therefore, the model remains agnostic to the specific details of the functions and the resource configurations. We tested our approach on well-known serverless systems like AWS lambda and Apache OpenWhisk and evaluated the system using 99 functional workloads. These workloads encompass both individual functions and chains of functions, addressing a range of computational and learning problems. To validate the system’s ability to adapt to changes, we also evaluated our system using functions with different input parameter sizes. Our evaluation shows that the model reaches convergence within 2 feedback iterations and results in a 32.41% decrease in average cost and a 5.18% average speedup, outperforming other state-of-the-art approaches.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100662"},"PeriodicalIF":4.5,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921259","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

Machine learning-based prediction of maternal continuum of care completion: Evidence from Bangladesh Demographic and Health Survey 2022 基于机器学习的孕产妇连续护理完成预测：来自孟加拉国2022年人口与健康调查的证据

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-26 DOI: 10.1016/j.array.2025.100666

Syed Toukir Ahmed Noor , Raisha Binte Islam , Samin Yeasar , Sazid Siddique

Despite decades of investments in maternal healthcare, the completion of the maternal continuum of care (CoC), defined as receiving at least four antenatal care visits with a medically trained provider, delivery by a skilled birth attendant at a health facility, and postnatal care within 48 h, remains low in Bangladesh. This study aimed to develop a machine learning-based prediction model to identify women most likely to complete the maternal CoC, using data from the 2022 Bangladesh Demographic and Health Survey. The analysis included 5,128 ever-married women aged 15–49 who had a live birth in the five years preceding the survey. After data pre-processing, including handling class imbalance through Synthetic Minority Oversampling Technique (SMOTE), 16 predictor variables were selected using the Boruta feature selection algorithm. Seven supervised machine learning algorithms were developed and evaluated, including Random Forest, XGBoost, support vector machines, logistic regression, and artificial neural networks. Among the 5,128 women, only 25.3 % completed the maternal CoC in Bangladesh. The random forest performed best with an accuracy of 82 %, precision of 81 %, recall of 71 %, F1-score of 0.758 and an AUC of 0.889. Feature importance and SHAP analysis identified wealth index, husband's and respondent's education, media exposure, and women's decision-making autonomy at household as the most influential predictors. Identifying women at high risk of dropout can enable healthcare providers to deliver targeted counseling and interventions. These findings offer valuable insights to inform data-driven policies and strategies to enhance maternal health service utilization and reduce preventable maternal morbidity and mortality in low-resource settings like Bangladesh.

尽管在孕产妇保健方面进行了数十年的投资，但在孟加拉国，孕产妇连续护理（CoC）的完成率仍然很低，其定义是由受过医学培训的提供者进行至少四次产前护理，在保健设施由熟练助产士接生，并在48小时内进行产后护理。本研究旨在利用2022年孟加拉国人口与健康调查的数据，开发一种基于机器学习的预测模型，以确定最有可能完成孕产妇CoC的妇女。该研究分析了5128名年龄在15岁至49岁之间的已婚女性，这些女性在调查前的五年内有过一次活产。数据预处理后，通过合成少数派过采样技术（Synthetic Minority Oversampling Technique， SMOTE）处理类失衡，使用Boruta特征选择算法选择16个预测变量。开发并评估了七种监督机器学习算法，包括随机森林、XGBoost、支持向量机、逻辑回归和人工神经网络。在孟加拉国的5128名妇女中，只有25.3%的人完成了孕产妇CoC。随机森林的准确率为82%，精密度为81%，召回率为71%，f1得分为0.758，AUC为0.889。特征重要性和SHAP分析发现，财富指数、丈夫和受访者的教育程度、媒体曝光率和女性在家庭中的决策自主权是最具影响力的预测因素。确定有辍学高风险的妇女可以使医疗保健提供者提供有针对性的咨询和干预措施。这些发现提供了宝贵的见解，为数据驱动的政策和战略提供信息，以便在孟加拉国等资源匮乏的环境中提高孕产妇保健服务的利用率，降低可预防的孕产妇发病率和死亡率。

{"title":"Machine learning-based prediction of maternal continuum of care completion: Evidence from Bangladesh Demographic and Health Survey 2022","authors":"Syed Toukir Ahmed Noor , Raisha Binte Islam , Samin Yeasar , Sazid Siddique","doi":"10.1016/j.array.2025.100666","DOIUrl":"10.1016/j.array.2025.100666","url":null,"abstract":"<div><div>Despite decades of investments in maternal healthcare, the completion of the maternal continuum of care (CoC), defined as receiving at least four antenatal care visits with a medically trained provider, delivery by a skilled birth attendant at a health facility, and postnatal care within 48 h, remains low in Bangladesh. This study aimed to develop a machine learning-based prediction model to identify women most likely to complete the maternal CoC, using data from the 2022 Bangladesh Demographic and Health Survey. The analysis included 5,128 ever-married women aged 15–49 who had a live birth in the five years preceding the survey. After data pre-processing, including handling class imbalance through Synthetic Minority Oversampling Technique (SMOTE), 16 predictor variables were selected using the Boruta feature selection algorithm. Seven supervised machine learning algorithms were developed and evaluated, including Random Forest, XGBoost, support vector machines, logistic regression, and artificial neural networks. Among the 5,128 women, only 25.3 % completed the maternal CoC in Bangladesh. The random forest performed best with an accuracy of 82 %, precision of 81 %, recall of 71 %, F1-score of 0.758 and an AUC of 0.889. Feature importance and SHAP analysis identified wealth index, husband's and respondent's education, media exposure, and women's decision-making autonomy at household as the most influential predictors. Identifying women at high risk of dropout can enable healthcare providers to deliver targeted counseling and interventions. These findings offer valuable insights to inform data-driven policies and strategies to enhance maternal health service utilization and reduce preventable maternal morbidity and mortality in low-resource settings like Bangladesh.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100666"},"PeriodicalIF":4.5,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921780","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

Research on different automatic segmentation methods for color cascading framework in detecting malaria infection 不同颜色级联框架自动分割方法在疟疾感染检测中的研究

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2025-12-23 DOI: 10.1016/j.array.2025.100658

Cucun Very Angkoso , Yonathan Ferry Hendrawan , Ari Kusumaningsih , Achmad Bauravindah , Rima Tri Wahyuningrum , Deshinta Arrova Dewi

Malaria remains a global health challenge due to its significant mortality and morbidity worldwide. Our study presents a comparative analysis of five automated segmentation models such as Fuzzy C-Means (FCM), K-Means, Gaussian Mixture Model (GMM), Entropy Filtering, and Watershed, for detecting malaria infection in microscopic blood smear images. The study uses dataset of 574 images, consisting of four Plasmodium species.

The study uses dataset of 574 images, comprising four Plasmodium species (P. falciparum, P. malariae, P. ovale, and P. vivax) across three developmental stages: trophozoites, schizonts, and gametocytes. Preprocessing techniques used is namely Color Cascading Method, including RGB normalization, gamma correction, noise reduction, exposure compensation, and edge enhancement, were employed to optimize image quality prior to segmentation. Experimental results demonstrate that FCM consistently outperforms other methods, achieving an average accuracy of 98.26 %, specificity of 97.91 %, and sensitivity of 98.61 %. The robustness of FCM in detecting faint and overlapping structures highlights its suitability for automated malaria diagnosis systems, particularly in resource-limited settings. While Entropy Filtering shows promise as an alternative, its inconsistent performance across different lifecycle stages limits its practical utility. The findings of this study provide a foundation for developing accurate and accessible automated diagnostic tools to enhance malaria detection and support global eradication efforts.

由于疟疾在世界范围内的死亡率和发病率很高，它仍然是一项全球健康挑战。本研究对模糊c均值（FCM）、k均值（K-Means）、高斯混合模型（GMM）、熵滤波（Entropy Filtering）和分水岭（Watershed）五种自动分割模型进行了比较分析，用于检测显微镜下血液涂片图像中的疟疾感染。该研究使用了574张图像的数据集，包括四种疟原虫。该研究使用了574张图像的数据集，包括四种疟原虫（恶性疟原虫、疟疾疟原虫、卵形疟原虫和间日疟原虫），它们跨越了三个发育阶段：滋养体、分裂体和配子体。使用的预处理技术即颜色级联法，包括RGB归一化、伽马校正、降噪、曝光补偿和边缘增强，在分割前优化图像质量。实验结果表明，FCM的平均准确率为98.26%，特异性为97.91%，灵敏度为98.61%，优于其他方法。FCM在检测微弱和重叠结构方面的鲁棒性突出了它对疟疾自动诊断系统的适用性，特别是在资源有限的环境中。虽然熵过滤作为一种替代方法很有希望，但它在不同生命周期阶段的不一致性能限制了它的实际应用。本研究结果为开发准确和可获得的自动化诊断工具提供了基础，以加强疟疾检测并支持全球根除工作。

{"title":"Research on different automatic segmentation methods for color cascading framework in detecting malaria infection","authors":"Cucun Very Angkoso , Yonathan Ferry Hendrawan , Ari Kusumaningsih , Achmad Bauravindah , Rima Tri Wahyuningrum , Deshinta Arrova Dewi","doi":"10.1016/j.array.2025.100658","DOIUrl":"10.1016/j.array.2025.100658","url":null,"abstract":"<div><div>Malaria remains a global health challenge due to its significant mortality and morbidity worldwide. Our study presents a comparative analysis of five automated segmentation models such as Fuzzy C-Means (FCM), K-Means, Gaussian Mixture Model (GMM), Entropy Filtering, and Watershed, for detecting malaria infection in microscopic blood smear images. The study uses dataset of 574 images, consisting of four Plasmodium species.</div><div>The study uses dataset of 574 images, comprising four Plasmodium species (P. falciparum, P. malariae, <em>P. ovale</em>, and <em>P. vivax</em>) across three developmental stages: trophozoites, schizonts, and gametocytes. Preprocessing techniques used is namely Color Cascading Method, including RGB normalization, gamma correction, noise reduction, exposure compensation, and edge enhancement, were employed to optimize image quality prior to segmentation. Experimental results demonstrate that FCM consistently outperforms other methods, achieving an average accuracy of 98.26 %, specificity of 97.91 %, and sensitivity of 98.61 %. The robustness of FCM in detecting faint and overlapping structures highlights its suitability for automated malaria diagnosis systems, particularly in resource-limited settings. While Entropy Filtering shows promise as an alternative, its inconsistent performance across different lifecycle stages limits its practical utility. The findings of this study provide a foundation for developing accurate and accessible automated diagnostic tools to enhance malaria detection and support global eradication efforts.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100658"},"PeriodicalIF":4.5,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921258","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