检索结果-内蒙古大学图书馆

multi-objective segmentation approach for bone age assessment using parameter tuning-based U-net architecture

multiMEDIA TOOLS AND APPLICATIONS 2022年第5期81卷 6755-6800页

作者： Deshmukh, Sonal Khaparde, Arti Maharashtra Inst Technol Dept Elect & Telecommun Pune Maharashtra India Dr Vishwanath Karadmit World Peace Univ Sch Elect & Commun Engn Pune Maharashtra India

Bone age assessment investigates the ossification improvement for estimating the skeletal age of the pediatrics for analyzing their skeletal growth and forecast their future adult height. The main intent of this paper is to contribute a novel deep learning-based bone segmentation for bone age assessment. Here, the datasets are gathered from both the manual as well as the RSNA database. The segmentation of 5 regions "Distal Phalanx of thumb, middle phalanx, third metacarpal, radius, and ulna" is performed by the optimized U-Net model. As an improvement in the existing U-Net architecture, tuning of the activation function is adopted by the hybridization of two meta-heuristic algorithms such as Class Topper Optimization (CTO) and Whale Optimization Algorithm (WOA) termed as Whale-based Class Topper Optimization (W-CTO). This improved model is developed with the intention of solving the multi-objective segmentation that concerns the parameters like entropy and variance. Moreover, the effect of the proposed segmentation is analyzed by estimating the bone age with the deep Convolutional Neural Network (Deep CNN). From the analysis, the overall MASE of W-CTO-U-Net+CNN is 14.66%, 22.06%, and 5.53% higher than RNN, CNN, and NN, respectively, and RMSE of W-CTO-U-Net+CNN is 53.28%, 22.02%, and 32.87% better than RNN, CNN, and NN, respectively.. The performance comparison of the proposed segmentation model over the conventional approaches confirms its effective performance with relatively high accuracy.

关键词： Bone age assessment multi-objective segmentation Optimized U-net architecture Whale-based class topper optimization Deep convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

MOTC: Abdominal multi-objective segmentation Model with Parallel Fusion of Global and Local Information

引用

JOURNAL OF IMAGING INFORMATICS IN MEDICINE 2024年第3期37卷 13-16页

作者： Zhang, GuoDong Gu, WenWen Wang, SuRan Li, YanLin Zhao, DaZhe Liang, TingYu Gong, ZhaoXuan Ju, RongHui Shenyang Aerosp Univ Sch Comp Daoyi South St Shenyang 110135 Liaoning Peoples R China Peoples Hosp Liaoning Prov Dept Radiol Wenyi St Shenyang 110016 Liaoning Peoples R China Northeastern Univ Key Lab Intelligent Comp Med Image Wenhua St Shenyang 110819 Liaoning Peoples R China

Convolutional Neural Networks have been widely applied in medical image segmentation. However, the existence of local inductive bias in convolutional operations restricts the modeling of long-term dependencies. The introduction of Transformer enables the modeling of long-term dependencies and partially eliminates the local inductive bias in convolutional operations, thereby improving the accuracy of tasks such as segmentation and classification. Researchers have proposed various hybrid structures combining Transformer and Convolutional Neural Networks. One strategy is to stack Transformer blocks and convolutional blocks to concentrate on eliminating the accumulated local bias of convolutional operations. Another strategy is to nest convolutional blocks and Transformer blocks to eliminate bias within each nested block. However, due to the granularity of bias elimination operations, these two strategies cannot fully exploit the potential of Transformer. In this paper, a parallel hybrid model is proposed for segmentation, which includes a Transformer branch and a Convolutional Neural Network branch in encoder. After parallel feature extraction, inter-layer information fusion and exchange of complementary information are performed between the two branches, simultaneously extracting local and global features while eliminating the local bias generated by convolutional operations within the current layer. A pure convolutional operation is used in decoder to obtain final segmentation results. To validate the impact of the granularity of bias elimination operations on the effectiveness of local bias elimination, the experiments in this paper were conducted on Flare21 dataset and Amos22 dataset. The average Dice coefficient reached 92.65% on Flare21 dataset, and 91.61% on Amos22 dataset, surpassing comparative methods. The experimental results demonstrate that smaller granularity of bias elimination operations leads to better performance.

关键词： multi-objective segmentation Parallel hybrid fusion Inter-Layer bias eliminate Abdominal CT image

来源：评论

学校读者我要写书评

暂无评论

Evolutionary multi-objective customer segmentation approach based on descriptive and predictive behaviour of customers: application to the banking sector

引用

JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE 2023年第8期35卷 1201-1223页

作者： Ben Ncir, Chiheb-Eddine Ben Mzoughia, Mohamed Qaffas, Alaa Bouaguel, Waad Univ Jeddah Coll Business Jeddah Saudi Arabia Univ Tunis ISG Tunis LARODEC Lab Tunis Tunisia Teamwill Consulting Grp Paris France

Customer segmentation is a challenging task in marketing that aims to build homogeneous segments of customers based on their similar characteristics and activities. This problem is considered multi-objective since it requires the evaluation of several variables including descriptive and predictive characteristics of customers. However, given that most exiting segmentation methods are based on the optimisation of a single-objective function, the identification of homogeneous customer segments in terms of both predictive and descriptive variables becomes a major issue. Descriptive and predictive characteristics are usually considered as two different and independent objectives, which cannot be optimised together. To deal with this problem, we propose a multi-objective segmentation approach based on three conceptual axes: descriptive, predictive, and quality-validation. In addition to the specificity of design of the multi-objective model, our proposed approach has the specificity of directly optimising the multi-objective problem using a customised genetic algorithm that directly approximates a set of Pareto-optimal solutions. We have applied and evaluated the proposed approach in an empirical study which aims to segment bank credit card customers using their descriptive characteristics and their predictive behaviour. Obtained results have shown the ability of the proposed approach to look for effective homogeneous segments and help decision-makers propose more tailored marketing strategies.

关键词： Unsupervised learning customer segmentation multi-objective segmentation genetic algorithm evolutionary algorithms customer lifetime value predicted customer behaviour

来源：评论

学校读者我要写书评

暂无评论

TransDiffSeg: Transformer-Based Conditional Diffusion segmentation Model for Abdominal multi-objective

引用

JOURNAL OF IMAGING INFORMATICS IN MEDICINE 2025年第1期38卷 262-280页

作者： Gu, WenWen Zhang, GuoDong Ju, RongHui Wang, SuRan Li, YanLin Liang, TingYu Guo, Wei Gong, ZhaoXuan Shenyang Aerosp Univ Sch Comp Daoyi South St Shenyang 110135 Liaoning Peoples R China Peoples Hosp Liaoning Prov Dept Radiol Wenyi St Shenyang 110016 Liaoning Peoples R China

In the domain of medical image segmentation, traditional diffusion probabilistic models are hindered by local inductive biases stemming from convolutional operations, constraining their ability to model long-term dependencies and leading to inaccurate mask generation. Conversely, Transformer offers a remedy by obviating the local inductive biases inherent in convolutional operations, thereby enhancing segmentation precision. Currently, the integration of Transformer and convolution operations mainly occurs in two forms: nesting and stacking. However, both methods address the bias elimination at a relatively large granularity, failing to fully leverage the advantages of both approaches. To address this, this paper proposes a conditional diffusion segmentation model named TransDiffSeg, which combines Transformer with convolution operations from traditional diffusion models in a parallel manner. This approach eliminates the accumulated local inductive bias of convolution operations at a finer granularity within each layer. Additionally, an adaptive feature fusion block is employed to merge conditional semantic features and noise features, enhancing global semantic information and reducing the Transformer's sensitivity to noise features. To validate the impact of granularity in bias elimination on performance and the impact of Transformer in alleviating the accumulated local inductive biases of convolutional operations in diffusion probabilistic models, experiments are conducted on the AMOS22 dataset and BTCV dataset. Experimental results demonstrate that eliminating local inductive bias at a finer granularity significantly improves the segmentation performance of diffusion probabilistic models. Furthermore, the results confirm that the finer the granularity of bias elimination, the better the segmentation performance.

关键词： Abdominal CT images multi-objective segmentation Diffusion probabilistic model Inductive bias elimination granularity Parallel combination

来源：评论

学校读者我要写书评

暂无评论

A method for accurate extraction of three-dimensional point cloud feature data in loading test of container ships

引用

OCEAN ENGINEERING 2024年 312卷

作者： Li, Rui Wan, Hexin Wang, Ji Huo, Shilin Soares, C. Guedes Dalian Univ Technol Sch Naval Architecture & Ocean Engn Dalian 116024 Liaoning Peoples R China Dalian Key Lab Adv Shipbldg Technol Dalian 116024 Liaoning Peoples R China Dalian Univ Technol State Key Lab Struct Anal Optimizat & CAE Software Dalian 116024 Liaoning Peoples R China Univ Lisbon Ctr Marine Technol & Ocean Engn CENTEC Inst Super Tecn Lisbon Portugal

The accuracy of container slot dimensions directly affects the efficiency and safety of container loading and unloading. However, traditional methods of container loading tests need to be improved for their long cycles, high costs, and low production efficiency. With the application of three-dimensional laser scanning devices in ship manufacturing, the use of simulated container loading test methods based on point cloud becomes a breakthrough in key technology for container ship construction. This paper presents a multi-objective segmentation method of cargo hold feature point cloud for container ship simulated loading tests. Data preprocessing algorithms are applied to remove noise and pseudo-image, and then a Random Sample Consensus algorithm is used to remove bulkhead and cargo bottom point clouds. Based on the density-based spatial clustering of applications with noise algorithm and the region growing algorithm, an algorithm for multiple segmentation of guide rail and bottom cone point clouds is devised, and a four-plane segmentation of guide rail point clouds is realized. Finally, a method for three-dimensional point cloud data flatness assessment and a Hausdorff distance based method for calculating the guide rail lateral and longitudinal spacing are proposed. These methods are validated through the loading test of a 16,000 TEU container ship.

关键词： Simulated container loading test Three-dimensional laser scanning Point cloud multi-objective segmentation Container ship construction

来源：评论

学校读者我要写书评

暂无评论

OF-AR Relation Aware Representation Learning for Lesion Image segmentation and Grading

OF-AR Relation Aware Representation Learning for Lesion Imag...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Yuan, Xinpan Jin, Siming Hua, Liujie Zhao, Guihu Zhang, Changhong Guo, Yuan College of Computer Science Hunan University of Technology Hunan ZhuZhou China Department of Neurology Xiangya Hospital Central South University Hunan Changsha China Department of Otolaryngology Zhuzhou People's Hospital Hunan Zhuzhou China

ISBN: (纸本)9798350368741

Medical image segmentation provides important supplementary information for lesion grading, but existing segmentation models usually only focus on the lesion area, which is susceptible to the influence of shooting distance and angle, leading to feature extraction errors. We found an "as one falls, another rises"(OF-AR) relationship between the lesion and the surrounding non lesion areas, and introduced OF-AR relationship aware learning representation to jointly extract features of lesions and non lesions. By dividing the image into black and white dual zones, expanding the feature extraction area, and using a dual zone contrastive learning module to increase the inter-class distance, accurate grading is ensured by utilizing the relative information of two complete targets. In addition, using weighted fusion methods to enhance the comprehensiveness and objectivity of grading. The experiment used adenoids as an example to verify the high accuracy of this method. © 2025 IEEE.

关键词： Adenoid Hypertrophy Dependency Relationship Diagnostic Grading Model Medical Image segmentation multi-objective segmentation

来源：评论

学校读者我要写书评

暂无评论

A Method to Optimize Peak Load Regulation in Power Grid Considering HVDC 4

A Method to Optimize Peak Load Regulation in Power Grid Cons...

引用

4th International Conference on Intelligent Green Building and Smart Grid (IGBSG)

作者： Xing, Chao Xi, Xinze Liu, Mingqun Yunnan Power Grid Co Ltd Elect Power Res Inst Kunming Yunnan Peoples R China

ISBN: (纸本)9781728121482

Peak load regulation problems may be aroused when the HVDC link is connected to the power system. In this paper, the evaluation method of multi-objective segmentation is proposed based on the HVDC operation modes. And the comprehensive evaluation model is established by considering membership functions of peak load regulated surplus quantity in the peak load condition and the abandoned electric quantity in the low load condition. The model utilizes the comprehensive membership functions to evaluate the effect of HVDC link on the power grid and optimize the peak load regulation. The proposed approach is applied to the Yunnan power grid in 2017, with the access to the HVDC link from Yongren to Funing to analyze the peak load regulation.

关键词： High voltage direct current peak load regulated surplus quantity abandoned electric quantity membership function multi-objective segmentation

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：