检索结果-内蒙古大学图书馆

Artificial intelligence breakthrough in diagnosis, treatment, and prevention of colorectal cancer - A comprehensive review

引用

BIOMEDICAL SIGNAL processing AND CONTROL 2025年 101卷

作者： Kumar, Anil Aravind, Nikhil Gillani, Tayiba Kumar, Deepak Post Grad Inst Med Educ & Res PGIMER Dept Pharmacol Chandigarh 160012 India Govt Coll Univ Inst Ind Biotechnol IIB Lahore Pakistan Translat Heath Sci & Technol Inst THSTI Faridabad 121001 India

Objective: Colorectal cancer (CRC) accounts for a significant number of deaths around the globe, with 1.2 million expected deaths by 2030, making it one of the major concerns in public health. Current diagnostic methods often lack timeliness, precision and often face challenges such as false positives and false negatives, leading to delays in effective intervention. These challenges underscore the need for more effective approaches, which serve as the objective of this review. One revolutionised solution to counter this need is the use of artificial intelligence (AI). Methodology: Leveraging machine learning, deep neural networks, and robotic systems, AI enhances early detection and personalized treatment strategies. This review explores the role of AI, particularly convolutional neural networks (CNNs) and recurrent neural networks (RNNs), in predicting CRC risk, AI based diagnosis through histopathological image analysis, and personalizing treatment plans using reinforcement learning techniques. Additionally, AI-driven robotic systems like da Vinci system have enabled precise removal of adenomatous polyps, improving patient outcomes. Results and conclusion: Combining artificial intelligence with other diagnostic and screening procedures has transformative impact on colorectal cancer management, emphasising breakthroughs in early detection, genetic testing, precision classification, and surgical procedures. Especially CADe enhanced adenoma detection in studies with low Adenoma detection rate (ADR), also increases detection of number of inconspicuous polyps which is normally missed by endoscopists. Overall, using artificial intelligence in medicine brings us closer to democratizing expertise and enhancing health outcomes for diverse populations. AI has potential to enhance diagnostic accuracy, optimise treatment strategies, and overcome limitations in traditional approaches while also addressing ethical and legal considerations.

关键词： Colorectal cancer Artificial Intelligence Artificial intelligence in colonoscopy machine learning Deep learning Robotic surgery

来源：评论

学校读者我要写书评

暂无评论

Leveraging physics-informed neural networks in geotechnical earthquake engineering: An assessment on seismic site response analyses

引用

COMPUTERS AND GEOTECHNICS 2025年 182卷

作者： Liu, Chenying Macedo, Jorge Rodriguez, Alexander Georgia Inst Technol North Ave NW Atlanta GA 30332 USA Univ Michigan Ann Arbor MI 48109 USA

The primary objective of this study is to assess the potential of physics-informed neural networks (PINNs) for seismic site response analyses (SRA). PINNs constitute a novel computational paradigm that combines physical principles with data-driven methods to solve differential equations. Despite the growing exploration of machine learning and deep learning in geotechnical earthquake engineering, the integration of PINNs remains limited. The study first addresses key challenges in applying PINNs to SRAs. In particular, the broad range of frequencies in ground motion recordings, which complicates the training process, and neural network architectural issues are discussed. Fourier feature embedding, a relatively new technique in image processing, learning rate adjustment, a tailored training strategy, and a PINN architecture are proposed to address the identified challenges. The proposed framework is evaluated by comparing SRA results from the implemented PINN and traditional numerical techniques, considering different ground motions and soil systems. The results of the proposed PINN and numerical techniques are identical, highlighting the robustness of the proposed framework. The encouraging results suggest there is significant potential for PINNs in general geotechnical earthquake engineering applications, which is also discussed.

关键词： Site response Deep learning Physics-informed neural network Spectral bias Ground shaking Geotechnical earthquake engineering

来源：评论

学校读者我要写书评

暂无评论

Purification of Contaminated Convolutional neural networks via Robust Recovery: An Approach With Theoretical Guarantee in One-Hidden-Layer Case

引用

IEEE Journal on Selected Topics in Signal processing 2025年第3期19卷 507-519页

作者： Lu, Hanxiao Huang, Zeyu Wang, Ren Illinois Institute of Technology Trustworthy and Intelligent Machine Learning Research Lab Department of Electrical and Computer Engineering ChicagoIL60616 United States Illinois Institute of Technology Department of Electrical and Computer Engineering ChicagoIL60616 United States

Convolutional neural networks (CNNs), one of the key architectures of deep learning models, have achieved superior performance on many machine learning tasks such as image classification, video recognition, and power systems. Despite their success, CNNs can be easily contaminated by natural noises and artificially injected noises such as backdoor attacks. In this paper, we propose a robust recovery method to remove the noise from the potentially contaminated CNNs and provide an exact recovery guarantee on one-hidden-layer non-overlapping CNNs with the rectified linear unit (ReLU) activation function. Our theoretical results show that both CNNs' weights and biases can be exactly recovered under the overparameterization setting with some mild assumptions. The experimental results demonstrate the correctness of the proofs and the effectiveness of the method in both the synthetic environment and the practical neural network setting. Our results also indicate that the proposed method can be extended to multiple-layer CNNs and potentially serve as a defense strategy against backdoor attacks. © 2007-2012 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Microstructural Analysis of Ceramic Materials Through machine learning Techniques: A Benchmarking Interdisciplinary Review for Ceramic Material Characterization

引用

Journal of The Institution of Engineers (India): Series D 2025年 1-19页

作者： Roy, Mahuya Kumar, Bijendra Chowdhury, Soumit Bhattacharya, Tapas Kumar Maity, Satyabrata Haldar, Partha Department of Computer Science and Engineering St. Thomas College of Engineering and Technology West Bengal Kolkata700023 India Gun and Shell Factory West Bengal Cossipore Kolkata700002 India Department of Computer Science and Engineering Jalpaiguri Government Engineering College West Bengal Jalpaiguri735102 India Department of Ceramic Technology Government College of Engineering and Ceramic Technology West Bengal Kolkata700010 India Department of Information Technology Techno International New Town 1/1 Service Rd. DG Block New Town Kolkata700156 India Department of Mechanical Engineering Government College of Engineering and Ceramic Technology West Bengal Kolkata700010 India

This study presents a revolutionary analytical study on ceramic microstructural images in the context of sophisticated image processing as well as machine learning techniques for automated and most accurate characterization of ceramic material properties. The main objective here is to scientifically correlate the derived microstructural image features with the concern material properties and this can be purely governed by specific material compositions that lead to better productivity. So, this particular unique study purely focuses on some efficient machine learning algorithms like K-means and fuzzy C-means in contrast to the traditional image processing concepts like Canny and Sobel for most effective classification as well as automated evaluation of ceramic microstructural image features. In addition, more advanced deep learning techniques like convolutional neural network also reflected here with in depth comparative analysis in view of its superiority for crack estimation, pore characterization, phase identification, grain boundary detection and grain area calculation that leads to accurate material characterizations. Further, continuous up gradations are still highlighted here with recent advanced deep learning models like Auto encoders, generative adversarial networks and recurrent neural networks for much more accurate characterization of ceramic microstructural image features boosting both material design and productivity in great manner. Apart from that, some latest image property related transformations are also investigated here for accurately tracing the critical image features while this study even adopts some exhaustive bibliometric analysis to precisely identify the effective microstructural evaluation methods from the existing literature. Hence this novel study not only provides a concrete pathway for identifying the most effective mechanism for ceramic microstructural image property evaluation and characterization in the present scenario but it als

关键词： Ceramic materials

来源：评论

学校读者我要写书评

暂无评论

Fusion strategies for deep convolutional neural network representations in histopathological image classification

引用

JOURNAL OF SUPERCOMPUTING 2025年第2期81卷 1-27页

作者： Osmani, Nooshin Esmaeeli, Erfan Rezayi, Sorayya Tarbiat Modares Univ Fac Med Sci Dept Med Informat Med Informat Tehran Iran Univ Tehran Med Sci Sch Allied Med Sci Dept Hlth Informat Management & Med Informat Hlth Informat Management Tehran Iran Aja Univ Med Sci Sch Paramed Sci Dept Hlth Informat Management Med Informat Tehran Iran

The enormous rise in processing power and the refinement of analysis algorithms over the past decade have resulted in the creation of several ways of computer-aided medical data analysis. We investigated the merits of combining a set of image classification systems incorporating 16 pre-trained convolutional neural networks with classifiers fusion strategy. We suggested a fusion paradigm that works on the posterior probability of different classifiers based on deep representations. For this purpose, motivated by the scarcity of medical images for training deep networks, a transfer learning paradigm was adopted where different deep convolutional networks (pre-trained on non-medical images) were used as mechanisms for feature extraction. Two well-known datasets including the animal diagnostics laboratory (ADL) and the breast cancer histopathological image classification (BreakHis) datasets were used. In the ADL dataset, the best result from the combination of classifiers trained on two pre-trained networks was 100% accuracy belong to the lung data. In the BreaKHis dataset, the suggested technique was investigated on both the image and patient levels which got the highest percentages of image classification rates compared to prior methods. This approach at 400 magnification achieved 92.7% at the image level, respectively, and 93.2% at the patient level. The Fisher's least significant difference post hoc test results showed a significant difference in accuracy parameters across the three classifiers reported in this research in the ADL and BreaKHis dataset with a p value < 0.001. The proposed classifier fusion strategy meets the best-reported performance for classifying histopathological images on two different histopathological datasets, ADL, and BreaKHis.

关键词： Histopathological images Deep convolutional networks Transfer learning Classifier fusion

来源：评论

学校读者我要写书评

暂无评论

A Survey on Automatic image Captioning Approaches: Contemporary Trends and Future Perspectives

引用

ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING 2025年第3期32卷 1459-1497页

作者： Salgotra, Garima Abrol, Pawanesh Selwal, Arvind Univ Jammu Dept Comp Sci & Informat Technol Jammu 180006 India Cent Univ Jammu Dept Comp Sci & Informat Technol Samba 181143 India

The automatic generation of image captions is one of the complex computer vision tasks that involve integration of object detection and natural language processing (NLP). In recent times, one of the significant aspects is to design image captioning approaches that accurately and efficiently generate appropriate image captions in a particular domain. With the emergence of deep learning paradigms, the task of image captioning becomes comparatively easier than traditional template-based approaches. In this article, we expound an in-depth examination of state of the art (SOTA) image captioning methods, along with the key conceptions. Besides, a comparative analysis of evaluation protocols is presented that are presently used to access the efficacy of the algorithms. Moreover, the study reveals open research issues in the existing methods that can be further investigated by the research community. One of the key challenges is to develop larger corpora of language specific dataset to design image captioning approaches in other regional languages such as Hindi, Marathi, Sanskrit, Telugu, and Gujarati etc. Furthermore, designing accurate and efficient image captioning approaches requisite the notion of attention mechanism in the images.

关键词： image captioning Text generation Deep learning machine translation neural networks

来源：评论

学校读者我要写书评

暂无评论

Application of machine learning algorithms in Artificial Intelligence image Recognition Technology 1

Application of Machine Learning Algorithms in Artificial Int...

引用

1st International conference on Intelligent systems and Computational networks, ICISCN 2025

作者： Ouyang, Qi Dai, Siyang Xie, Kexin Sun, Zeyu Wang, Yao Polytechnic of Milan Milan Italy Faculty of Engineering and Information Technology The University of Melbourne VIC Australia College of Business and Public Management Wenzhou-Kean University Wenzhou China School of Environment and Science Griffith University Brisbane Australia Nanjing Stomatological Hospital Affiliated Hospital of Medical School Institute of Stomatology Nanjing University Nanjing China

ISBN: (纸本)9798331529246

In smart city applications such as medical image analysis, autonomous driving, and security monitoring, image recognition faces challenges like complex backgrounds, low-quality images, and diverse targets, affecting accuracy and robustness. This study explores machine learning algorithms to enhance image recognition performance by addressing noise interference, intricate backgrounds, and feature extraction difficulties. It combines Convolutional neural networks (CNN) with transfer learning, starting with data preprocessing to reduce noise, using pre-trained CNN models to extract high-level features, and fine-tuning with a ResNet transfer learning strategy for specific tasks. Additionally, ensemble learning methods are employed to further improve model robustness and accuracy. Experimental results show that the ensemble model maintains around 85% accuracy even with high background complexity, and transfer learning achieves 90% accuracy when the sample size reaches 1000. These findings demonstrate that transfer and ensemble learning effectively enhance image recognition accuracy and resilience in complex environments. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Mean-Square Synchronization of Additive Time-Varying Delayed Markovian Jumping neural networks Under Multiple Stochastic Sampling

引用

IEEE TRANSACTIONS ON neural networks AND learning systems 2025年 PP卷 PP页

作者： Anbalagan, Pratap Feng, Zhiguang Huang, Tingwen Cui, Yukang Shenzhen Univ Coll Mechatron & Control Engn Shenzhen 518060 Peoples R China Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen 518060 Peoples R China Harbin Engn Univ Coll Automat Harbin 150001 Peoples R China Shenzhen Univ Adv Technol Fac Comp Sci & Control Engn Shenzhen 518055 Peoples R China Peng Cheng Lab Shenzhen 518000 Peoples R China

This study aims to solve the mean-square asymptotic synchronization problem of additive time-varying delayed Markovian jumping neural networks (ATVMJNNs) under the framework of multiple stochastic samplings and its direct application in secure image encryption (SIE). To do this, first, we assume the existence of multiple sampled data periods that satisfy a Bernoulli distribution and introduce random variables to represent the positions of input delays and sampling periods. Then, based on these assumptions, we develop a mode-dependent discontinuous Lyapunov-Krasovskii functional (DLKF) to reduce model conservatism. Next, we introduce a new auxiliary slack-matrix-based integral inequality (ASMBII) to approximate the integral quadratic terms arising from the derivative of the DLKFs. Furthermore, we develop a multiple stochastic sampling framework to achieve asymptotic synchronization between the primary and secondary systems, and less conservative criteria for asymptotic stability in the mean square sense of the error model are derived by solving a set of linear matrix inequalities (LMIs). Finally, we present the numerical validations and corresponding experimental results in a pragmatic application of image processing to demonstrate the benefits of the proposed algorithms and techniques. From both numerical and practical results, the proposed algorithms and techniques can yield superior performance compared to existing studies.

关键词： Synchronization Stochastic processes Artificial neural networks Delays Symmetric matrices Delay effects Additives Linear matrix inequalities Fluctuations Encryption Additive time-varying delays (ATDs) discontinuous Lyapunov functional Markovian jump parameters mean-square asymptotic synchronization multiple stochastic sampling secure image encryptions (SIEs)

来源：评论

学校读者我要写书评

暂无评论

State-of-the-art techniques for optical underwater image enhancement

引用

INTERNATIONAL JOURNAL OF image AND DATA FUSION 2025年第1期16卷

作者： Joseph, Naveena Tresa Kumar, S. N. Suriyan, Kannadhasan Amal Jyothi Coll Engn Dept CSE APJAKTU Koovappally India Amal Jyothi Coll Engn Dept EEE Koovappally 686518 Kerala India Study World Coll Engn Dept ECE Coimbatore India

Underwater image enhancement is an area of greater concern as underwater images present several hurdles due to the light propagation properties in aquatic environments. Critical application in areas of marine biology, offshore inspection, and underwater archaeology needs crystal clear images for fact-finding and inferring results. This review article comprises a comprehensive overview of methods for enhancing underwater images ranging from traditional techniques to advanced deep-learning approaches. Histogram-based approaches, filtering techniques, colour correction methods, and image fusion-based approaches are examined under conventional methods. In contrast, deep learning-based approaches including Convolutional neural networks (CNN) and Generative Adversarial networks (GAN) produced potential results generating realistic underwater images. Firstly, we discuss the various pre-processing steps used in underwater image enhancement, succeeded by various noise reduction and image enhancement methods. A thorough and impartial analysis of existing literature is not currently available because of variations in the experimental setup and dataset used. Therefore, we analyse typical image enhancement algorithms on the HICRD dataset and evaluate them quantitatively and qualitatively based on the standard evaluation metrics. Utilizing the insights from classical and cutting-edge technologies, this review helps researchers select appropriate techniques for enhancing underwater imagery.

关键词： Underwater image enhancement colour correction image fusion machine learning (ML) deep learning (DL) Convolutional neural Network (CNN) Generative Adversarial Network (GAN)

来源：评论

学校读者我要写书评

暂无评论

Deep learning for Multimodal Breast Cancer Characterization With Emergence of Terahertz and Infrared Imaging

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Gezimati, Mavis Singh, Ghanshyam Univ Johannesburg Ctr Smart Informat & Commun Syst Dept Elect & Elect Engn Sci Auckland Pk Kingsway ZA-2006 Johannesburg South Africa

While significant progress has been made in multimodal image data fusion tasks, the application of deep learning for multisource processing of terahertz (THz) and infrared (IR) image datasets has not yet been explored. It is through deep learning-based multisource data fusion and computational modeling that outcomes of interest, such as smart clinical decision support systems, ultimately facilitate improved patient-level decision-making. In this article, we are focusing on multisource medical image data processing streamlined to the disease classification and characterization tasks. Specifically, we propose the framework and algorithms for multimodal deep learning-based models for classifying THz and IR pixel multimodal datasets for breast cancer applications. Initially, a brief overview of the principles of multimodal data fusion, the dynamics and interaction mechanisms in biological systems, as well as parameter extraction and modeling in both THz and IR thermography is reported. A deep learning-based multimodal framework for breast cancer classification and characterization in IR thermography and THz imaging is proposed. Further, for proof of concept, the decision-level fusion is performed on the IR thermography dataset to test the model performance. The model performance is evaluated using the metrics which are achieved as: validation accuracy of 96.6% and sensitivity of 96.3%. The confusion matrix and receiver operating characteristics are computed with the area under the curve (AUC) of 0.99. Last, a roadmap for future research is proposed based on recent trends.

关键词： Terahertz radiation Imaging Terahertz communications Cancer Deep learning Breast cancer Biomedical imaging Spectroscopy Infrared imaging Data integration convolutional neural networks (CNNs) detection diagnosis infrared (IR) thermal imaging multimodal framework terahertz (THz) imaging

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：