检索结果-内蒙古大学图书馆

A pyramid convolutional mixer for cervical pap-smear image classification tasks

BIOMEDICAL signal processing AND CONTROL 2025年 99卷

作者： Yang, Tianjin Hu, Hexuan Li, Xing Qing, Meng Chen, Linhai Huang, Qian Hohai Univ Coll Comp & Software Nanjing 211100 Peoples R China Nanjing Forestry Univ Coll informat Sci & Technol Nanjing 210037 Peoples R China Nanjing Forestry Univ Coll Artificial Intelligence Nanjing 210037 Peoples R China Univ Hong Kong Dept Real Estate & Construct Hong Kong Peoples R China

Convolutional neural Networks (CNNs) have exhibited considerable success in the realm of cervical cytopathology image classification, owing to their efficient design. We find that existing CNN-based cervical cytopathology classification methods fail to fully exploit the cell morphology and nucleus information. To address the above problems, we propose an efficient network called Pyramid Convolutional Mixer. We capture multi-scale subtle morphology features at the cellular level and convey nuclear neighborhood spatial information by integrating convolutional operations within the transformer structure. PCMixer contains two key modules, i.e. pyramid morphology module (PMM) and nuclear spatial mixing block (NSMB) to retrieve cervical cytopathology information. PMM is characterized by a multi-scale pyramid architecture employing a convolutional layer and a local encoder to generate local morphology information at each scale. In addition, NSMB operates on the input patches to separate the mixing of spatial and channel dimensions to encode nuclear neighborhood spatial information. We intend to unveil a more intricate cervical cytopathology dataset: Cervical Cytopathology image Dataset (CCID). We achieve a classification accuracy of 89.62% along with precision, recall and F1 score of 82.76%, 85.97% and 84.15% respectively on the CCID dataset. Also, we use cervical cytopathology images from the publicly available SIPaKMeD dataset. We obtain 96.21%, 95.70% 95.60% and 95.30% respectively for the four metrics. Through comprehensive experiments conducted on two real-world datasets, our proposed model demonstrates superior performance compared to state-of-the-art cervical cytopathology classification models. The results demonstrate that our method can significantly assist cytopathologists in appropriately evaluating cervical smears.

关键词： Large inner-class variance Subtle inter-class variance Cervical cytopathology image classification Pyramid convolutional mixer

来源：评论

学校读者我要写书评

暂无评论

Data-Augmented Numerical Integration in State Prediction: Rule Selection 20

Data-Augmented Numerical Integration in State Prediction: Ru...

引用

20th IFAC Symposium on System Identification (SYSID)

作者： Dunik, J. Kral, L. Matousek, J. Straka, O. Brandner, M. Univ West Bohemia Dept Cybernet Univ 8 Plzen 30601 Czech Republic Univ West Bohemia Dept Math Univ 8 Plzen 30601 Czech Republic

This paper deals with the state prediction of nonlinear stochastic dynamic systems. The emphasis is laid on a solution to the integral Chapman-Kolmogorov equation by a deterministic-integration-rule-based point-mass method. A novel concept of reliable data-augmented, i.e., mathematics- and data-informed, integration rule is developed to enhance the point-mass state predictor, where the trained neural network (representing data contribution) is used for the selection of the best integration rule from a set of available rules (representing mathematics contribution). The proposed approach combining the best properties of the standard mathematics-informed and novel data-informed rules is thoroughly discussed. Copyright (c) 2024 The Authors.

关键词： State estimation neural network Numerical integration Nonlinear predictors Bayesian methods stochastic systems

来源：评论

学校读者我要写书评

暂无评论

State space models meet transformers for hyperspectral image classification

引用

signal processing 2025年 226卷

作者： Shi, Xuefei Zhang, Yisi Liu, Kecheng Wen, Zhaokun Wang, Wenxuan Zhang, Tianxiang Li, Jiangyun Univ Sci & Technol Beijing Sch Automat & Elect Engn Beijing 100083 Peoples R China Univ Sci & Technol Beijing Shunde Grad Sch Foshan 528000 Peoples R China Minist Educ Key Lab Knowledge Automat Ind Proc Beijing 100083 Peoples R China

In recent years, convolutional neural networks and vision transformers have emerged as predominant models for hyperspectral remote sensing image classification task, leveraging staked convolution layers and self- attention mechanisms with high computation costs, respectively. Recent studies, such as the Mamba model, have showcased the ability of state space model (SSM) with efficient hardware-aware designs in efficiently modeling sequences and extracting implicit features along tokens, which is precisely needed for accurate hyperspectral image (HSI) classification. Thus making SSM-based model potentially a new architecture for remote sensing HSI classification task. However, SSM encounters challenges in modeling HSI due to the insensitivity of spatial information and redundant spectral characteristics. Given SSM-based methods rarely explored in HSI classification, in this work, we present the first exploration of SSM-based models for HSI classification task. Our proposed method MamTrans effectively leverages the capacity of transformer for capturing spatial tokens relationships and Mamba for extracting implicit features along tokens. Besides, we propose a Bidirectional Mamba Module to enhance SSM's spatial perception ability of extracting spatial features in HSI. Our proposed MamTrans obtains a new state-of-the-art performance across five commonly employed HSI classification benchmarks, demonstrating the robust generalization of MamTrans and effectiveness of SSM-based structure for HSI classification task. Our codes could be found at https://***/PPPPPsanG/MamTrans.

关键词： Hyperspectral image classification Remote sensing State space model Mamba Transformer

来源：评论

学校读者我要写书评

暂无评论

Blind, all-depth dispersion compensation and blind spectral calibration 28

Blind, all-depth dispersion compensation and blind spectral ...

引用

Conference on Optical Coherence Tomography and Coherence Domain Optical methods in Biomedicine XXVIII

作者： Maliszewski, Krzysztof A. Urbanska, Magdalena A. Vetrova, Varvara Kolenderska, Sylwia M. Univ Canterbury Sch Math & Stat Christchurch New Zealand Massey Univ Sch Food & Adv Technol Palmerston North New Zealand Univ Canterbury Sch Phys & Chem Sci Christchurch New Zealand Nicolaus Copernicus Univ Fac Phys Astron & Informat Grudziadzka 5 PL-87100 Torun Poland

ISBN: (纸本)9781510669208;9781510669192

We present a neural network able to fully linearise an OCT image without any a priori knowledge about the spectrometer characteristics or the extent of dispersion in the interferometer and the object. Unlike the earlier solutions, this blind linearisation is not biased towards a specific object, nor its dispersion characteristics, and in the future can be made independent of the light source parameters.

关键词： Dispersion neural networks signal to noise ratio Spectral calibration image processing Optical coherence tomography

来源：评论

学校读者我要写书评

暂无评论

INTERPRETING CONVOLUTIONAL neural NETWORKS BY EXPLAINING THEIR PREDICTIONS 30

INTERPRETING CONVOLUTIONAL NEURAL NETWORKS BY EXPLAINING THE...

引用

30th IEEE International Conference on image processing (ICIP)

作者： Meynen, Toon Behzadi-Khormouji, Hamed Oramas, Jose Univ Antwerp Imec IDLab Antwerp Belgium

ISBN: (纸本)9781728198354

We propose a method that exploits the feedback provided by visual explanation methods combined with pattern mining techniques to identify the relevant class-specific and class-shared internal units. In addition, we put forward a patch extraction approach to find faithfully class-specific and class-shared visual patterns. Contrary to the common practice in literature, our approach does not require pushing augmented visual patches through the model. Experiments on two CNN architectures show the effectiveness of the proposed method.

关键词： neural Network Interpretation Data Mining

来源：评论

学校读者我要写书评

暂无评论

neural nonnegative matrix factorization for hierarchical multilayer topic modeling

引用

SAMPLING THEORY signal processing AND DATA ANALYSIS 2024年第1期22卷 1-38页

作者： Haddock, Jamie Will, Tyler Vendrow, Joshua Zhang, Runyu Molitor, Denali Needell, Deanna Gao, Mengdi Sadovnik, Eli Harvey Mudd Coll Dept Math 301 Platt Blvd Claremont CA 91711 USA Optimal Dynam New York NY 10001 USA MIT Dept EECS 50 Vassar St Cambridge MA 02140 USA Harvard Univ Sch Engn & Appl Sci 150 Western Ave Cambridge MA 02138 USA Google Seattle WA 98103 USA Univ Calif Los Angeles Dept Math 520 Portola Plaza Los Angeles CA 90095 USA Schlumberger Menlo Pk CA 94025 USA

We introduce a new method based on nonnegative matrix factorization, neural NMF, for detecting latent hierarchical structure in data. Datasets with hierarchical structure arise in a wide variety of fields, such as document classification, image processing, and bioinformatics. neural NMF recursively applies NMF in layers to discover overarching topics encompassing the lower-level features. We derive a backpropagation optimization scheme that allows us to frame hierarchical NMF as a neural network. We test neural NMF on a synthetic hierarchical dataset, the 20 Newsgroups dataset, and the MyLymeData symptoms dataset. Numerical results demonstrate that neural NMF outperforms other hierarchical NMF methods on these data sets and offers better learned hierarchical structure and interpretability of topics.

关键词： Hierarchical topic models Nonnegative matrix factorization Backpropagation

来源：评论

学校读者我要写书评

暂无评论

Input parameterized physics informed neural networks for de noising, super-resolution, and imaging artifact mitigation in time resolved three dimensional phase-contrast magnetic resonance imaging

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 150卷

作者： Kalajahi, Amin Pashaei Csala, Hunor Bin Mamun, Zayeed Yadav, Sangeeta Amili, Omid Arzani, Amirhossein D'Souza, Roshan M. Univ Wisconsin Milwaukee Dept Mech Engn Milwaukee WI 53211 USA Univ Utah Dept Mech Engn 1495 E 100 S Salt Lake City UT 84112 USA Univ Utah Sci Comp & Imaging Inst 72 S Cent Campus Dr Ste 3750 Salt Lake City UT 84112 USA Univ Toledo Mech Ind & Mfg Engn Dept 2801 West Bancroft St Toldeo OH 43606 USA

Motivation: Hemodynamic analysis is crucial for diagnosing and predicting cardiovascular diseases. However, methods relying on fluid flow simulations or blood flow imaging are complex, time-consuming, and require specialized expertise, limiting their clinical use. Goal: This research aims to automate the enhancement of blood flow images, providing clinicians with a fast, accurate tool for hemodynamic analysis without requiring advanced expertise. Objectives: A software tool based on physics-constrained neural networks was developed to enable clinicians to easily select and process regions of interest (ROIs) in time-resolved three-dimensional phase contrast magnetic resonance imaging (4D-Flow MRI) blood flow images for quick, accurate analysis. methods: The Input Parameterized Physics-Informed neural Network (IP-PINN) was introduced to improve the spatio-temporal resolution of 4D-Flow MRI. IP-PINN mitigates noise, velocity aliasing, and phase errors. A convolutional neural network processes ROI data into latent vectors, which are then used to predict velocity, pressure, and spin density via a multi-layer perceptron. The method is trained with synthetic blood flow data using an innovative loss function that addresses noise and artifacts. Results: IP-PINN successfully enhanced image resolution, reducing noise and artifacts when tested on synthetic 4D-Flow MRI data derived from blood flow simulations of intracranial aneurysms. For data with 20 decibels (dB) signal-to-noise ratio, results closely matched the ground truth with less than 5.5% relative error. processing took under two minutes. The method also has the potential to reduce data acquisition time by 25%. Conclusions: IP-PINN could significantly enhance the clinical use of 4D-Flow MRI for personalized hemodynamic analysis in cardiovascular diseases.

关键词： Phase contrast magnetic resonance imaging Super-resolution Generalizability of physics-informed neural nets

来源：评论

学校读者我要写书评

暂无评论

HAU-Net: Hybrid CNN-transformer for breast ultrasound image segmentation

引用

BIOMEDICAL signal processing AND CONTROL 2024年第PartB期87卷

作者： Zhang, Huaikun Lian, Jing Yi, Zetong Wu, Ruichao Lu, Xiangyu Ma, Pei Ma, Yide Lanzhou Univ Sch Informat Sci & Engn Lanzhou Gansu Peoples R China Lanzhou Jiaotong Univ Sch Elect & Informat Engn Lanzhou Gansu Peoples R China

Breast cancer is a significant health concern that remains one of the leading causes of mortality in women worldwide. Convolutional neural Networks (CNNs) have been shown to be effective in ultrasound breast image segmentation. Yet, because of the lack of long-distance dependence, the segmentation performance of CNNs is limited in addressing challenges typical of segmentation of ultrasound breast lesions, such as similar intensity distributions, the presence of irregular objects, and blurred boundaries. In order to overcome these issues, several studies have combined transformers and CNNs, to compensate for the shortcomings of CNNs with the ability of transformers to exploit long-distance dependence. Most of these studies limited themselves to rigidly plug transformer blocks into the CNN, lacking consistency in the process of feature extraction and therefore leading to poor performances in segmenting challenging medical images. In this paper, we propose HAU-Net(hierarchical attention-guided U-Net), a hybrid CNN-transformer framework that benefits from both the long-range dependency of transformers and the local detail representation of CNNs. To incorporate global context information, we introduce a L-G transformer block nested into the skip connections of the U shape architecture network. In addition, to further improve the segmentation performance, we added a cross attention block (CAB) module on the decoder side to allow different layers to interact. Extensive experimental results on three public datasets indicate that the proposed HAU-Net can achieve better performance than other state-of-the-art methods for breast lesions segmentation, with Dice coefficient of 83.11% for BUSI, 88.73% for UDIAT, and 89.48% for BLUI respectively.

关键词： Convolutional neural Network Transformer Breast lesions segmentation Ultrasound image

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Based Non-Invasive Framework for Neurodegenerative Parkinson's Disease Diagnosis from Template-Less Handwritten Shapes 4

A Deep Learning Based Non-Invasive Framework for Neurodegene...

引用

4th International Conference on Robotics, Electrical and signal processing Techniques, ICREST 2025

作者： Nawal, Rifa Habib, Nafisha Barua, Sancoy Department of Eee Chattogram4349 Bangladesh

ISBN: (纸本)9798331530938

Parkinson's disease (PD) is a neurodegenerative disorder affecting millions globally. Current diagnostic methods lack precision. In this research, an approach combining Histogram of Oriented Gradients (HOG) with a custom Convolutional neural Network (CNN) for early PD diagnosis is proposed. A medical image dataset was curated, and preprocessing techniques were employed for image enhancement and uniformity. The CNN extracts key features while HOG provides edge orientation information, and their fusion creates a robust feature map. An integrated attention mechanism further refines focus on crucial regions. Evaluation demonstrates a balanced performance in terms of accuracy (99%) and parameter (0.8M) requirement. Visualization tools like GradCAM offer insights into model decisions, aiding interpretability. This approach offers an accurate PD detection, potentially transforming diagnosis and improving patient outcomes. Code will be made available at https://***/rifanawal99/P-DL. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Universal BlackMarks: Key-image-Free Blackbox Multi-Bit Watermarking of Deep neural Networks

引用

IEEE signal processing LETTERS 2023年 30卷 36-40页

作者： Li, Li Zhang, Weiming Barni, Mauro Shanghai Univ Sch Commun & Informat Engn Shanghai 200444 Peoples R China Univ Sci & Technol China CAS Key Lab Electromagnet Space Informat Hefei 230026 Peoples R China Univ Siena Departmen Informat Engn & Math Via Roma I-53100 Siena Italy

Existing methods for Deep neural Networks (DNN) watermarking either require accessing the internal parameters of the DNN models (white-box watermarking), or rely on backdooring to enforce a desired behavior of the model when the DNN is fed with a specific set of key input images (black-box watermarking). In this letter, we propose a black-box multi-bit DNN watermarking algorithm, suitable for multiclass classification networks, whereby the presence of the watermark can be retrieved from the output of the network in correspondence to any input. To read the watermark, we first apply a power function to the softmax output of the DNN model to map it from an impulse-like to a smooth distibution. Then, we extract the watermark bits by projecting the output of the DNN onto a pseudorandom key vector. Watermark embedding is achieved by adding a proper regularizer term to the training loss. The effectiveness of the proposed method is demonstrated by applying it to various network architectures working on different datasets. The experimental results demonstrate the possibility to embed a robust watermark into the output of the host DNN with a negligible impact on the accuracy of the original task.

关键词： Watermarking Closed box Training Decoding Payloads Task analysis neural networks DNN Watermarking Blackbox watermarking multi-bit watermarking key-image-free watermarking

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：