检索结果-内蒙古大学图书馆

Evaluating masked self-supervised learning frameworks for 3D dental model segmentation tasks

Scientific Reports 2025年第1期15卷 1-15页

作者： Krenmayr, Lucas von Schwerin, Reinhold Schaudt, Daniel Riedel, Pascal Hafner, Alexander Geserick, Marc Cooperative Doctoral Program for Data Science and Analytics University of Ulm Ulm 89081 Germany Department of Computer Science University of Applied Sciences Ulm 89081 Germany smyl tp GmbH Ulm 89073 Germany

The application of deep learning using dental models is crucial for automated computer-aided treatment planning. However, developing highly accurate models requires a substantial amount of accurately labeled data. Obtaining this data is challenging, especially in the medical domain. Masked self-supervised learning has shown great promise in overcoming the challenge of data scarcity. However, its effectiveness has not been well explored in the 3D domain, particularly on dental models. In this work, we investigate the applicability of the four recently published masked self-supervised learning frameworks-Point-BERT, Point-MAE, Point-GPT, and Point-M2AE-for improving downstream tasks such as tooth and brace segmentation. These frameworks were pre-trained on a proprietary dataset of over 4000 unlabeled 3D dental models and fine-tuned using the publicly available Teeth3DS dataset for tooth segmentation and a self-constructed braces segmentation dataset. Through a set of experiments we demonstrate that pre-training can enhance the performance of downstream tasks, especially when training data is scarce or imbalanced—a critical factor for clinical usability. Our results show that the benefits are most noticeable when training data is limited but diminish as more labeled data becomes available, providing insights into when and how this technique should be applied to maximize its effectiveness. © The Author(s) 2025.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimization of the Use of Artificial Neural Network Models for Accuracy Data Measurement Palm Oil Production Prediction Rate

Optimization of the Use of Artificial Neural Network Models ...

引用

2023 International Conference on computer science, Information Technology and Engineering, ICCoSITE 2023

作者： Setiawan, Iwan R. Fanani, Ahmad Z. Surname, Given Name Purwanto, Purwanto Dian Nuswantoro University Doctoral Program in Computer Science Semarang Indonesia

ISBN: (纸本)9798350320954

Predictions are necessary in modern times if the amount of production, profit and loss analysis, and development must be calculated on a specific scale. The average prediction only determines a certain period using data series, so it will be very helpful on certain production scales, especially palm oil production as a commodity with a very large amount of investment. Palm oil production is Indonesia's production potential, which requires a prediction pattern for a certain period. The autoregressive integrated moving average (ARIMA) model is used to calculate production rates by performing predictive data analysis of palm oil production development over a period of time. To prove the level of foresight in the prediction process, the next step is to calculate the level of foresight of the prediction process error through the Mean Square Error (MSE) model. To get the accuracy value, the Artificial Neural Network (ANN) algorithm approach is carried out using the MATLAB tool. There are several ways to test the level of accuracy;the results are compared with the calculation results using MS Excel as well as the Statistical Product and Service Solution (SPSS) tool. The calculated accuracy of SPSS (0.015) and MS Excel (0.0201) are known, while the ANN algorithm is more observant with a value of 0.0055. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Comparison of Effectiveness of Logistic Regression, Naive Bayes, and Random Forest Algorithms in Predicting Student Arguments 7

Comparison of Effectiveness of Logistic Regression, Naive Ba...

引用

7th Information Systems International Conference, ISICO 2023

作者： Wahyuningsih, Tri Manongga, Danny Sembiring, Irwan Wijono, Sutarto Doctoral Program of Computer Science Universitas Kristen Satya Wacana Salatiga Indonesia

Currently, in the process of assessing and giving feedback on students' argumentative writing, educators have to spend a considerable amount of time reading and analyzing each essay individually. This can be a complicated and time-consuming process, especially if the number of students to be assessed is quite large. The problem of this research is to find the most effective algorithm in providing accurate and reliable predictions in the context of evaluation and feedback of students' argumentation. This study compares three algorithms (logistic regression, Naive Bayes, and Random Forest) to predict student argumentation using essays from grades 6-12. Logistic regression performed best with 94.34% accuracy, followed by random forest with 91.98% accuracy, and Naive Bayes with 88.93% accuracy. The study optimized preprocessing and selected algorithms for an automated guidance model. It is the first stage of a three-part study for developing automated guidance models. Data came from Kaggle, and the study aims to improve the accuracy of automated guidance models for student argumentation. © 2023 The Authors. Published by Elsevier B.V.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Cray/Blitz-1984 chess champion

引用

Telematics and Informatics 1985年第4期2卷 299-305页

作者： Hyatt, Robert Nelson, Harry Gower, Albert Mr. Hyatt is in the doctoral program in the Department of Computer Science the University of Alabama at Birmingham United States

The chess program "Cray Blitz" is the current World computer Chess Champion and North American computer Chess Champion (ACM). The program has also played in human chess tournaments and is a chess master. At speed chess, where its ability to perform very accurate analysis is particularly important, it has maintained a performance rating of over 2600 for the past two years. This indicates that at speed chess, the program is one of the top players, electronic or human, in the world. It is currently running on a Cray XMP-48 computer system and has been designed around the parallelism that the XMP architecture provides. © 1986.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic Model for Determining Disaster Evacuation Locations with Game Theory 1

Dynamic Model for Determining Disaster Evacuation Locations ...

引用

1st International Conference on Advanced Information Scientific Development, ICAISD 2020

作者： Safii, M. Efendi, Syahril Zarlis, Muhammad Mawengkang, Herman Doctoral Program Faculty of Computer Science and Information Technology Universitas Sumatera Utara Indonesia

Disaster management can be grouped into three stages with several activities that can be carried out starting from pre-disaster, during emergency response and post-disaster. In the pre-disasterstage one of which is mitigation, which is a series of efforts to reduce disaster risk, both through physical development and awareness raising and capacity to face the threat of disaster. In mitigation one important factor is about evacuation. Evacuation locations in disaster transportation management are urgently needed for an effective evacuation process. Evacuation location determination is a benchmark in handling evacuation that can minimize the impact of losses from the disaster. To determine the evacuation location dynamically using game theory can solve the problem by providing several alternative locations by showing the dominant factor in the variables in the selection of the evacuation site. Game theory will use pure and mixed strategies to provide the most optimum choice of criteria. With this method will produce a decision support system to determine the optimal evacuation location or safe area in disaster mitigation © 2020 Published under licence by IOP Publishing Ltd.

关键词： Site selection

来源：评论

学校读者我要写书评

暂无评论

Enhancing Diagnostic Accuracy for Knee Osteoarthritis Using a Deep Learning Approach

Enhancing Diagnostic Accuracy for Knee Osteoarthritis Using ...

引用

International Symposium on Applied science 2024, ISAS 2024

作者： Raza, Aquib Ching, Congo Tak Shing Liu, Peng-Ta Phan, Thien-Luan Tran, Ngoc-Luan Nguyen, Ha Anh T. Rameez, Mohammad Li, Hung-Chung Linh, Dao Nguyen Phuong Van Hieu, Nguyen Tran, Trung Nghia National Taiwan University Taipei City10673 Taiwan Research Center for Information Technology Academia Sinica Taipei City11529 Taiwan Graduate Institute of Biomedical Engineering National Chung Hsing University Taichung402 Taiwan Department of Electrical Engineering National Chi Nan University Nantou County Taiwan Advanced Plant and Food Crop Biotechnology Center National Chung Hsing University Taichung402 Taiwan International Doctoral Program in Agriculture National Chung Hsing University Taichung402 Taiwan Fall Prevention Center Department of Physical Medicine & Rehabilitation Changhua Christian Hospital Changhua500 Taiwan Department of Physics and Electronic Engineering University of Science Vietnam National University - Ho Chi Minh City Ho Chi Minh City70000 Viet Nam Department of Computer Science University of York York HeslingtonYO10 5DD United Kingdom Undergraduate Program of Intellectual Creativity Engineering National Chung Hsing University Taichung402 Taiwan Ho Chi Minh City72409 Viet Nam HCMUT Vietnam National University Ho Chi Minh City Thu Duc Ho Chi Minh City71308 Viet Nam

This research introduces a deep learning framework that combines convolutional neural networks with autoencoders to improve the diagnostic accuracy of knee osteoarthritis. The study utilized a publicly available dataset from the Mendeley Data repository, where knee joint images were preprocessed using Contrast Limited Adaptive Histogram Equalization to enhance contrast and ensure consistent image sizing. Initially, an autoencoder was employed in an unsupervised manner to learn compressed representations of the data. These representations were then used to train a CNN to classify knee images into 'healthy' and 'osteoarthritic' categories. The model was trained using an 80:20 split for training and testing data, incorporating early stopping and data augmentation techniques to prevent overfitting and enhance generalization. Performance metrics such as accuracy, precision, recall, and the area under the Receiver Operating Characteristic curve (ROCAUC) were used to evaluate the model, resulting in a high test accuracy of 98.95%. The high precision and recall rates further confirmed the model's robustness in diagnosing KOA. To enhance interpretability, Gradient-weighted Class Activation Mapping (Grad-CAM) was applied to identify critical regions in the images that influenced the model's decisions. Additionally, t-Distributed Stochastic Neighbor Embedding (t-SNE) was used to visualize the high-dimensional encoded features in a two-dimensional space, providing deeper insights into how the model differentiates between classes. The promising results of this deep learning approach suggest its potential for broader applications in medical imaging diagnostics. By providing healthcare professionals with a tool that enhances diagnostic accuracy and operational efficiency, this framework lays the foundation for developing scalable, AI-driven diagnostic systems for various complex medical conditions. © 2025 Institute of Physics Publishing. All rights reserved.

关键词： Joints (anatomy)

来源：评论

学校读者我要写书评

暂无评论

CARe: An ontology for representing context of activity-aware healthcare environments

CARe: An ontology for representing context of activity-aware...

引用

2011 AAAI Workshop

作者： Rodríguez, Marcela D. Tentori, Mónica Favela, Jesús Saldaña, Diana García, Juan-Pablo Master and Doctoral Program of Computer Science Autonomous University of Baja California Mexico Computer Science Department CICESE Research Center Mexico

ISBN: (纸本)9781577355205

Representing computational activities is still an open problem in the field of Activity-Aware Computing. In this paper, drawn from our experiences çdeveloping activity-aware applications in support of two populations: nurses working in hospitals and elders living independently;we defined the Context Aware Representational (CARe) model. CARe is an ontology that enables the representation and management of computational activities. We illustrate, through application scenarios, that the CARe ontology is flexible enough to enable developers to customize the computational representation of healthcare activities. Copyright © 2011, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Impact Analysis of RGB Channels to the Quality of Imperceptibility in Image Steganography 5

Impact Analysis of RGB Channels to the Quality of Impercepti...

引用

5th International Conference on Information and Communications Technology, ICOIACT 2022

作者： Ignatius Moses Setiadi, De Rosal Rustad, Supriadi Andono, Pulung Nurtantio Shidik, Guruh Fajar Soeleman, M. Arief Pujiono Purwanto Dian Nuswantoro University Computer Science Doctoral Program Semarang Indonesia Dian Nuswantoro University Computer Science Faculty Semarang Indonesia

ISBN: (数字)9781665451406

ISBN: (纸本)9781665451406

In computers, color images generally have three color channels, namely red (R), green (G), and blue (B). Embedding a message in the image can change the pixel value. This change certainly adjusts the image pixel value, luminance, contrast, structure, and even image characteristics. So, it is necessary to know how much effect the pixel changes have on each channel. By learning how the effects occur, future research can find solutions to minimize the impact of changes. Based on testing of 40 standard image datasets with four different steganographic methods. The red channel has the highest sensitivity to embedding based on MSE, PSNR, and SSIM measurements. The green has better SSIM, and the blue channel has better PSNR and MSE. However, overall, the difference of MSE, PSNR, dan SSIM from the three channels is less significant © 2022 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Self-referentiality in the Brouwer-Heyting-Kolmogorov semantics of intuitionistic logic

Self-referentiality in the Brouwer-Heyting-Kolmogorov semant...

引用

International Symposium on Logical Foundations of computer science, LFCS 2013

作者： Yu, Junhua Doctoral Program in Computer Science Graduate Center City University of New York United States

ISBN: (纸本)9783642357213

The intended provability semantics of Intuitionistic Propositional Logic IPC (also called Brouwer-Heyting-Kolmogorov semantics) has been formalized within Gödel-Artemov's framework. According to this approach, IPC is embedded in modal logic S4 by Gödel embedding and S4 is realized in Artemov's Logic of Proofs LP which has a provability interpretation in Peano Arithmetic. Artemov's realization of S4 in LP uses self-referential LP-formulas of the form t : ö(t), namely, 't is a proof of a formula φ containing t itself.' Kuznets showed that this is not avoidable by offering S4-theorems that cannot be realized without using self-referential LP-formulas. This paper extends Kuznets' method to find IPC-theorems that call for direct self-referentiality in LP. Roughly speaking, examples include double-negations of tautologies that are not IPC-theorems, e.g., ≠≠(≠≠p → p), and there are also examples in the purely implicational fragment IPC →. This suggests that the Brouwer-Heyting-Kolmogorov semantics of intuitionistic logic is intrinsically self-referential. © Springer-Verlag 2013.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Efficient Deep Learning Model in Road Surface Condition Monitoring

Efficient Deep Learning Model in Road Surface Condition Moni...

引用

2022 International Seminar on Application for Technology of Information and Communication, iSemantic 2022

作者： Robet, Robet Hasibuan, Zainal Arifin Arief Soeleman, Mochammad Purwanto, Purwanto Andono, Pulung Nurtantio Pujiono, Pujiono Dian Nuswantoro University Computer Science Doctoral Program Semarang Indonesia Dian Nuswantoro University Computer Science Faculty Semarang Indonesia

ISBN: (纸本)9781665488396

Road is a crucial asset for public services. Therefore, it must be repaired as soon as possible to bring safety for the driver on the road. In recent years research in monitoring and identifying road surface conditions using images captured by the low-cost camera or smartphone cameras and deep learning can identify road surface conditions efficiently and quickly. The study aims to explore the performance of the deep learning model. We proposed a U-Net architecture for road extraction from Road Traversing Knowledge (RTK) dataset. The result has been compared with different automatic segmentation models, and the development of segmentation using U-Net showed a more precise Accuracy of 97.08%, Mean-IoU 0.364, and faster computing time than the other models. © 2022 IEEE.

关键词： Cameras

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：