检索结果-内蒙古大学图书馆

Machine learning based small bowel video capsule endoscopy analysis: Challenges and opportunities

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 2023年 143卷 191-214页

作者： Wahab, Haroon Mehmood, Irfan Ugail, Hassan Sangaiah, Arun Kumar Muhammad, Khan Univ Bradford Ctr Visual Comp Bradford England Natl Yunlin Univ Sci & Technol Int Grad Inst Artificial Intelligence Touliu 64002 Taiwan Lebanese Amer Univ Dept Elect & Comp Engn Byblos Lebanon Sungkyunkwan Univ Coll Comp & Informat Sch Convergence Dept Appl Artificial IntelligenceVisual Analyt Kn Seoul 03063 South Korea

Video capsule endoscopy (VCE) is a revolutionary technology for the early diagnosis of gastric disorders. However, owing to the high redundancy and subtle manifestation of anomalies among thousands of frames, the manual construal of VCE videos requires considerable patience, focus, and time. The automatic analysis of these videos using computational methods is a challenge as the capsule is un-tamed in motion and captures frames inaptly. Several machine learning (ML) methods, including recent deep convolutional neural networks approaches, have been adopted after evaluating their potential of improving the VCE analysis. However, the clinical impact of these methods is yet to be investigated. This survey aimed to highlight the gaps between existing ML-based research methodologies and clinically significant rules recently established by gastroenterologists based on VCE. A framework for interpreting raw frames into contextually relevant frame-level findings and subsequently merging these findings with meta-data to obtain a disease-level diagnosis was formulated. Frame-level findings can be more intelligible for discriminative learning when organized in a taxonomical hierarchy. The proposed taxonomical hierarchy, which is formulated based on pathological and visual similarities, may yield better classification metrics by setting inference classes at a higher level than training classes. Mapping from the frame level to the disease level was structured in the form of a graph based on clinical relevance inspired by the recent international consensus developed by domain experts. Furthermore, existing methods for VCE summarization, classification, segmentation, detection, and localization were critically evaluated and compared based on aspects deemed significant by clinicians. Numerous studies pertain to single anomaly detection instead of a pragmatic approach in a clinical setting. The challenges and opportunities associated with VCE analysis were delineated. A focus on maxim

关键词： Machine learning visual computing Deep learning Capsule endoscopy Small bowel Precision diagnostic

来源：评论

学校读者我要写书评

暂无评论

A Systematic Review of Facial Expression Detection Methods

引用

IEEE ACCESS 2023年 11卷 61881-61891页

作者： Pinto, Lyanh Vinicios Lopes Alves, Andre Vinicius Neves Medeiros, Alana Miranda Costa, Saulo William da Silva Pires, Yomara Pinheiro Costa, Fernando Augusto Ribeiro Seruffo, Marcos Cesar da Rocha Fed Univ Para Fac Comp Engn & Telecommun BR-66075110 Belem Para Brazil Fed Univ Para Fac Comp Campus Castanhal BR-68746630 Castanhal Para Brazil Fed Univ Para Postgrad Program Elect Engn BR-66075110 Belem Para Brazil Fed Inst Para BR-68250000 Obidos Para Brazil Fed Univ Para Ctr High Amazon Studies BR- 66075110 Balem Para Brazil

Understanding emotions is one of the greatest capabilities of human beings, as it allows the understanding of facial expressions that facilitate the capture of important information about other individuals, which are used for the perception of mental or emotional states. Advances in Artificial Intelligence and visual computing, more specifically in Deep Learning with the advent of Artificial Neural Networks, have enhanced the ability of machines to infer human emotions through image analysis. This paper presents a Systematic Literature Review (SLR) with the purpose of researching, mapping and summarizing studies that address techniques or algorithms more efficiently. The convolutional neural network models analyzed in this review are based on deep learning with an emphasis on expression and microexpression recognition. The results suggest that database uses, with laboratory controlled images, combined with CNN's such as VGG and ResNet, have excellent performances in their tests. For better understanding, we will detail and compare all the methods obtained in the review.

关键词： Emotion recognition convolutional neural networks deep learning databases visual computing

来源：评论

学校读者我要写书评

暂无评论

From Graphical Treatment of Combinatorics to Tiling Grammars

引用

NEXUS NETWORK JOURNAL 2023年第SUPPL 1期25卷 321-332页

作者： Walter, Natalie Ligler, Heather Guersoy, Benay Penn State Univ Dept Architecture 121 Stuckeman Family Bldg University Pk PA 16802 USA

This research employs shape grammars to generate variations of the Truchet tile to characterize a two-dimensional and three-dimensional generative tiling system.

关键词： Shape grammars Tessellations tilings Design analysis Transformations visual computing

来源：评论

学校读者我要写书评

暂无评论

Towards Holistic Analysis of Wireless Capsule Endoscopic Videos: A Taxonomy-Driven Machine Learning Framework for Clinically Comprehensive WCE Frame Level Analysis 4

Towards Holistic Analysis of Wireless Capsule Endoscopic Vid...

引用

4th IEEE Interdisciplinary Conference on Electrics and Computer (INTCEC)

作者： Wahab, Haroon Goel, Rita Alamgir, Maida Mehmood, Irfan Muhammad, Khan Ugail, Hassan Univ Bradford Ctr Visual Comp & Intelligent Syst Sch Comp Sci AI & Elect Bradford W Yorkshire England Sungkyunkwan Univ Dept Appl Artificial Intelligence Visual Analyt Knowledge Lab VIS2KNOW Seoul 03063 South Korea

ISBN: (纸本)9798350349467;9798350349450

Manual perusal of capsule endoscopic videos is laborious due to their length, redundancy, and poor quality frames, risking misdetection. Limited representation of diverse anomalies and subtle lesions led to less generalized approaches targeting only individual or few anomalous findings. To deploy automated analysis in clinical settings effectively, a holistic, end-to-end approach is crucial, distinguishing between frame-level analysis and disease-level diagnosis. This study proposes a holistic framework for frame-level analysis based on a taxonomy of findings, leveraging clinical consensus. Clinical knowledge implicit in the taxonomy can enhance Computer Aided Diagnostics (CAD) systems. Convolutional Neural Networks (CNNs) trained at appropriate granularity levels in the taxonomy show improved performance metrics. Validation on the Kvasir-capsule dataset demonstrates a 2-4% performance boost. Taxonomy-based inference scores effectively resolve mimicked anomalies without additional learning.

关键词： Capsule Endoscopy Deep Learning Machine Learning Precision Diagnostics Small Bowel visual computing

来源：评论

学校读者我要写书评

暂无评论

Weakly Supervised Method for Domain Adaptation in Instance Segmentation 40th

Weakly Supervised Method for Domain Adaptation in Instance S...

引用

40th Computer Graphics International Conference (CGI)

作者： Sun, Jie Tian, Yan Wang, Jialei Xu, Zhaocheng Wang, Hao Jiang, Zhaoyi Wang, Xun Zhejiang Gongshang Univ Sch Comp Sci & Technol Hangzhou 310018 Peoples R China Shining 3D Tech Co Ltd Hangzhou 311258 Peoples R China Massey Univ Sch Math & Computat Sci Auckland 0632 New Zealand

ISBN: (纸本)9783031500688;9783031500695

The domain adaptation of an instance segmentation model has gained much attention. However, manual annotation is tedious and self-training contains too much pseudolabel noise. Inspired by weakly supervised methods, we propose a method to handle these challenges by limited verification signals and label propagation. Semantic trees are constructed to explore the relation between samples by using a clustering method;Then, reliable pseudolabels are verified and propagated to unreliable labels, which improves instance segmentation model by employing the updated samples. Experiments on public datasets demonstrate that the proposed approach is competitive with state-of-the-art approaches.

关键词： visual computing Deep Learning Instance Segmentation Domain Adaptation

来源：评论

学校读者我要写书评

暂无评论

House Price Prediction via visual Cues and Estate Attributes 17th

House Price Prediction via Visual Cues and Estate Attributes

引用

17th International Symposium on visual computing (ISVC)

作者： Vaddi, Sai Surya Yousif, Amira Baraheem, Samah Shen, Ju Nguyen, Tam, V Univ Dayton Dept Comp Sci Dayton OH 45469 USA

ISBN: (纸本)9783031207150;9783031207167

The price of a house depends on many factors, such as its size, location, amenities, surrounding establishments, and the season in which the house is being sold, just to name a few of them. As a seller, it is absolutely essential to price the property competitively else it will not attract any buyers. This problem has given rise to multiple companies as well as past research works that try to enhance the predictability of property prices using relevant mathematical models and machine learning techniques. In this research, we investigate the usage of machine learning in predicting the house price based on related estate attributes and visual images. To this end, we collect a dataset of 2,000 houses across different cities in the United States. For each house, we annotate 14 estate attributes and five visual images for exterior, interior-living room, kitchen, bedroom, and bathroom. Following the dataset collection, different features are extracted from the input data. Furthermore, a multi-kernel regression approach is used to predict the house price from both visual cues and estate attributes. The extensive experiments demonstrate the superiority of the proposed method over the baselines.

关键词： House price prediction Multi-kernel learning visual computing Estate attributes Computer vision Machine learning

来源：评论

学校读者我要写书评

暂无评论

Contrast and Aggregation Network for Generalized Zero-shot Learning 31st

Contrast and Aggregation Network for Generalized Zero-shot L...

引用

31st International Conference on Artificial Neural Networks (ICANN)

作者： Li, Bin Xie, Cheng Yang, Jingqi Duan, Haoran Yunnan Univ Kunming Yunnan Peoples R China

ISBN: (纸本)9783031159312;9783031159305

Generalized zero-shot learning (GZSL) aims to classify both seen classes and unseen classes by training seen classes and the semantic information shared by seen and unseen classes. Some methods try to synthesize unseen samples by unseen semantic information;however, the performance of the GZSL is limited as the distribution of synthesized samples and of real samples is misalignment. To tackle this problem, we propose a Contrast and Aggregation Network (CAN). CAN employs a feature and contrast module to align the real and synthesized feature, which maps the real and synthesized visual feature to a new space and aggregates both real and synthesized features to the centroid of real visual features. Since the semantic information may be lost during the generating and mapping stage, we designed a semantic feature alignment module to project the mapped feature into semantic space. We evaluate our proposed CAN on four benchmark datasets to demonstrate that our method is against the state-of-the-art GZSL methods. The source code is available at https://***/fesfa/CAN.

关键词： Zero-shot learning Image classification visual computing Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

From theory to practice: Truchet tile as a computational visual coding approach for facade visualization

引用

INTERNATIONAL JOURNAL OF ARCHITECTURAL computing 2024年

作者： EL-Mahdy, Deena British Univ Egypt Fac Engn Architectural Engn Dept Suez St Cairo 11837 Egypt

The study presents unique teaching approach for generating a parametric modular facade fabrication for first-year students in visual design module using a computational visual coding method based on Truchet Tiles. The method followed four main phases: exploring tiling, testing connectivity, fabrication, and assembly. Shape grammar was considered during the design principles implementation to respond to climatic behavior. Space-filling shapes method known as the "Generalized Abeille Tiles" was followed to ensure tiling connectivity. The experimentation involves manual folding techniques without any software. A survey was conducted among first-year and fourth-year students post-introduction to computational software. The results showed a wide range of alternative geometries with different self-shading patterns generated. It highlighted that facade fabrication aided students in comprehending computational logic, easing the understanding of computational software complexity. This pilot study emphasizes the effectiveness of teaching computational concepts in early curricula.

关键词： Truchet tile visual computing algorithm responsive fa & ccedil ade parametric design topology interlocking assembly

来源：评论

学校读者我要写书评

暂无评论

A language to analyze, describe, and explore collections of visual art

引用

visual computing for Industry,Biomedicine,and Art 2021年第1期4卷 29-49页

作者： Hermann Pflüger Institute for Visualization and Interactive Systems University of Stuttgart70569 StuttgartGermany

A vast quantity of art in existence today is inaccessible to *** people want to know the different types of art that exist,how individual works are connected,and how works of art are interpreted and discussed in the context of other works,they must utilize means other than simply viewing the ***,this paper proposes a language to analyze,describe,and explore collections of visual art(LadeCA).LadeCA combines human interpretation and automatic analyses of images,allowing users to assess collections of visual art without viewing every image in *** paper focuses on the lexical base of *** also outlines how collections of visual art can be analyzed,described,and explored using a LadeCA ***,the relationship between LadeCA and indexing systems,such as ICONCLASS or AAT,is demonstrated,and ways in which LadeCA and indexing systems can complement each other are highlighted.

关键词： Collections of visual art visual computing Personalized digital libraries Digital humanities Semantics

来源：评论

学校读者我要写书评

暂无评论

Exploration of overlap volumes for radiotherapy plan evaluation with the aim of healthy tissue sparing

引用

COMPUTERS IN BIOLOGY AND MEDICINE 2023年 166卷 107523-107523页

作者： Schlachter, Matthias Peters, Samuel Camenisch, Daniel Putora, Paul Martin Buehler, Katja Forsch GmbH VRVis Zent Virtual Real & Visualisierung Vienna Austria Kantonsspital St Gallen Dept Radiat Oncol St Gallen Switzerland Univ Bern Dept Radiat Oncol Bern Switzerland

Purpose: Development of a novel interactive visualization approach for the exploration of radiotherapy treatment plans with a focus on overlap volumes with the aim of healthy tissue sparing. Methods: We propose a visualization approach to include overlap volumes in the radiotherapy treatment plan evaluation process. Quantitative properties can be interactively explored to identify critical regions and used to steer the visualization for a detailed inspection of candidates. We evaluated our approach with a user study covering the individual visualizations and their interactions regarding helpfulness, comprehensibility, intuitiveness, decision-making and speed. Results: A user study with three domain experts was conducted using our software and evaluating five data sets each representing a different type of cancer and location by performing a set of tasks and filling out a questionnaire. The results show that the visualizations and interactions help to identify and evaluate overlap volumes according to their physical and dose properties. Furthermore, the task of finding dose hot spots can also benefit from our approach. Conclusions: The results indicate the potential to enhance the current treatment plan evaluation process in terms of healthy tissue sparing.

关键词： visualization visual computing Radiotherapy planning Healthy tissue sparing Overlap volume

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：