检索结果-内蒙古大学图书馆

1st Deep Breast Workshop on AI and Imaging for Diagnostic and Treatment Challenges in Breast Care, Deep-Breath 2024

作者： Tanyel, Toygar Denizoglu, Nurper Seker, Mustafa Ege Alis, Deniz Cerekci, Esma Karaarslan, Ercan Aribal, Erkin Oksuz, Ilkay Biomedical Engineering Graduate Program Istanbul Technical University Istanbul Turkey Department of Radiology Acibadem Healthcare Group Istanbul Turkey Department of Radiology University of Wisconsin-Madison MadisonWI United States Department of Radiology School of Medicine Acibadem University Istanbul Turkey Sisli Hamidiye Etfal Training and Research Hospital Istanbul Turkey Department of Computer Engineering Istanbul Technical University Istanbul Turkey

ISBN: (纸本)9783031777882

Breast cancer remains a leading cause of cancer-related deaths among women worldwide, with mammography screening as the most effective method for the early detection. Ensuring proper positioning in mammography is critical, as poor positioning can lead to diagnostic errors, increased patient stress, and higher costs due to recalls. Despite advancements in deep learning (DL) for breast cancer diagnostics, limited focus has been given to evaluating mammography positioning. This paper introduces a novel DL methodology to quantitatively assess mammogram positioning quality, specifically in mediolateral oblique (MLO) views using attention and coordinate convolution modules. Our method identifies key anatomical landmarks, such as the nipple and pectoralis muscle, and automatically draws a posterior nipple line (PNL), offering robust and inherently explainable alternative to well-known classification and regression-based approaches. We compare the performance of proposed methodology with various regression and classification-based models. The CoordAtt UNet model achieved the highest accuracy of 88.63% ± 2.84 and specificity of 90.25% ± 4.04, along with a noteworthy sensitivity of 86.04% ± 3.41. In landmark detection, the same model also recorded the lowest mean errors in key anatomical points and the smallest angular error of 2.42∘. Our results indicate that models incorporating attention mechanisms and CoordConv module increase the accuracy in classifying breast positioning quality and detecting anatomical landmarks. Furthermore, we make the labels and source codes available to the community to initiate an open research area for mammography, accessible at https://***/tanyelai/deep-breast-positioning. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

7-Level Soft Charging Switched Capacitor Multilevel Inverter

引用

IEEE Access 2025年

作者： Memis, Mehmet Karakilic, Murat Van Yüzüncü Yıl University Başkale Vocational School Electrical and Energy Program Van Turkey Igdir University Department of Computer Engineering Igdir Turkey

This paper proposes a novel 7-level switched capacitor multilevel inverter (SC-MLI) topology with soft charging capability and analyzes the performance in terms of efficiency and component optimization. The proposed structure achieves high efficiency up to 97.32% while producing the same output level with fewer components compared to existing SC-MLI designs in the literature. Moreover, the total cost function (CF) value of 3.26 is among the lowest values, indicating that the system offers an economically advantageous solution. Simulation and experimental studies confirm that the system generates stable output waveforms under resistive and inductive loads and operates stably over a wide frequency and modulation index (MI) range. In particular, the effect of the soft charging technique on capacitor currents has been investigated and it has been found that this technique minimizes switching losses by reducing inrush current transients. The SPWM method using high switching frequency produces a natural soft charging effect on the capacitor currents, producing lower peak currents compared to NLC control and improving the overall efficiency of the system. The study shows that SC-MLI is optimized as a function of low component count, high efficiency and low cost. The results reveal that this structure is suitable for high-performance applications such as renewable energy systems, electric vehicle power conversion systems, uninterruptible power supplies (UPS), microgrids, and motor drives. These findings provide an important reference for the future development of SC-MLI designs. © 2013 IEEE.

关键词： Uninterruptible power systems

来源：评论

学校读者我要写书评

暂无评论

BAQET: BRAM-aware Quantization for Efficient Transformer Inference via Stream-based Architecture on an FPGA 25

BAQET: BRAM-aware Quantization for Efficient Transformer Inf...

引用

Proceedings of the 2025 ACM/SIGDA International Symposium on Field programmable Gate Arrays

作者： Ling-Chi Yang Chi-Jui Chen Trung Le Bo-Cheng Lai Scott Hauck Shih-Chieh Hsu Institute of Electronics National Yang Ming Chiao Tung University Hsinchu Taiwan Graduate Degree Program of College of Electrical and Computer Engineering in National Yang Ming Chiao Tung University Hsinchu Taiwan Electrical and Computer Engineering in University of Washington Seattle USA Electrical and Computer Engineering University of Washington Seattle USA Physics University of Washington Seattle USA

ISBN: (纸本)9798400713965

FPGAs are a compelling substrate for supporting machine learning inference. Tools such as High-Level Synthesis and hls4ml can shorten the development cycle for deploying ML algorithms on FPGAs, but can struggle to handle the large on-chip storage needed for many of these models. In particular the high BRAM usage found in many of these flows can cause Place & Route failures during synthesis. In this paper we propose using a Simulated-Annealing based flow to perform BRAM-aware quantization. This approach trades off inference accuracy with BRAM usage, to provide a high-quality inference engine that still meets on-chip resource constraints. We demonstrate this flow for Transformer-based machine learning algorithms, which include Flash Attention in a Stream-based Dataflow architecture. Our system imposes minimal accuracy drops, yet can reduce BRAM usage by 20%-50%, and improve power efficiency by 264%-812% compared to existing Transformer-based accelerators on FPGAs

关键词： efficient transformer

来源：评论

学校读者我要写书评

暂无评论

ATP_mCNN: Predicting ATP binding sites through pretrained language models and multi-window neural networks

引用

computers in Biology and Medicine 2025年 185卷 109541-109541页

作者： Le, Van-The Malik, Muhammad-Shahid Lin, Yi-Jing Liu, Yu-Chen Chang, Yan-Yun Ou, Yu-Yen Department of Computer Science and Engineering Yuan Ze University Chung-Li32003 Taiwan Graduate Program in Biomedical Informatics Yuan Ze University Chung-Li32003 Taiwan Department of Computer Sciences Karakoram International University Gilgit-Baltistan 15100 Pakistan

Adenosine triphosphate plays a vital role in providing energy and enabling key cellular processes through interactions with binding proteins. The increasing amount of protein sequence data necessitates computational methods for identifying binding sites. However, experimental identification of adenosine triphosphate-binding residues remains challenging. To address the challenge, we developed a multi-window convolutional neural network architecture taking pre-trained protein language model embeddings as input features. In particular, multiple parallel convolutional layers scan for motifs localized to different window sizes. Max pooling extracts salient features concatenated across windows into a final multi-scale representation for residue-level classification. On benchmark datasets, our model achieves an area under the ROC curve of 0.95, significantly improving on prior sequence-based models and outperforming convolutional neural network baselines. This demonstrates the utility of pre-trained language models and multi-window convolutional neural networks for advanced sequence-based prediction of adenosine triphosphate-binding residues. Our approach provides a promising new direction for elucidating binding mechanisms and interactions from primary structure. © 2024 Elsevier Ltd

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Debiasing Classifiers by Amplifying Bias with Latent Diffusion and Large Language Models 25

Debiasing Classifiers by Amplifying Bias with Latent Diffusi...

引用

Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing

作者： Donggeun Ko Dongjun Lee Namjun Park Wonkyeong Shim Jaekwang Kim AiM Future Seoul Republic of Korea Maum AI Gyeonggi-do Republic of Korea Department of Computer Science and Engineering/Convergence Program for Social Innovation Sungkyunkwan University Suwon Republic of Korea Department of Computer Science and Engineering Sungkyunkwan University Suwon Republic of Korea School of Convergence / Convergence Program for Social Innovation Sungkyunkwan University Suwon Republic of Korea

ISBN: (纸本)9798400706295

Neural networks struggle with image classification when biases are learned and misleads correlations, affecting their generalization and performance. Previous methods require attribute labels (e.g. background, color) or utilizes Generative Adversarial Networks (GANs) to mitigate biases. We introduce DiffuBias, a novel pipeline for text-to-image generation that generates bias-conflict samples, without any training. By utilizing pretrained diffusion and image captioning models, DiffuBias generates, bias-conflict samples using the top-K losses from a biased classifier (fB) to debias the classifier. This method not only debiases effectively but also boosts classifier generalization capabilities. Our comprehensive experimental evaluations demonstrate that DiffuBias achieves state-of-the-art performance on benchmark datasets.

关键词： debiasing

来源：评论

学校读者我要写书评

暂无评论

MathReader : Text-to-Speech for Mathematical Documents

MathReader : Text-to-Speech for Mathematical Documents

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Sieun Hyeon Kyudan Jung Nam-Joon Kim Hyun Gon Ryu Jaeyoung Do Department of Electrical and Computer Engineering Seoul National University Department of Mathematics Chung-Ang University NVIDIA Interdisciplinary Program in Artificial Intelligence Seoul National University

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

TTS (Text-to-Speech) document reader from Microsoft, Adobe, Apple, and OpenAI have been serviced worldwide. They provide relatively good TTS results for general plain text, but sometimes skip contents or provide unsatisfactory results for mathematical expressions. This is because most modern academic papers are written in LaTeX, and when LaTeX formulas are compiled, they are rendered as distinctive text forms within the document. However, traditional TTS document readers output only the text as it is recognized, without considering the mathematical meaning of the formulas. To address this issue, we propose MathReader, which effectively integrates OCR, a fine-tuned T5 model, and TTS. MathReader demonstrated a lower Word Error Rate (WER) than existing TTS document readers, such as Microsoft Edge and Adobe Acrobat, when processing documents containing mathematical formulas. MathReader reduced the WER from 0.510 to 0.281 compared to Microsoft Edge, and from 0.617 to 0.281 compared to Adobe Acrobat. This will significantly contribute to alleviating the inconvenience faced by users who want to listen to documents, especially those who are visually impaired. The code is available at https://***/hyeonsieun/MathReader.

关键词： Text recognition Error analysis Pipelines Optical character recognition Graphics processing units Signal processing Real-time systems Mathematical models Text to speech Speech processing

来源：评论

学校读者我要写书评

暂无评论

Constructing Semantical Structure by Segmentation Integrated Video Embedding for Temporal Action Detection

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Zhao, Zixuan Liu, Shuming Zhao, Chengze Zhao, Xu Shanghai Jiao Tong University School of Electronic Information and Electrical Engineering Shanghai200240 China King Abdullah University of Science and Technology Computer Science Program Thuwal23599 Saudi Arabia

Video embedding is the pivot in Temporal Action Detection (TAD). Once the video embedding can robustly capture the essence of actions and perceive activities in complex scenes, the TAD model can more accurately localize action boundaries. Currently, video embedding is typically based on rule-based pixel convolution or cube-based transformer, wherein structured semantic information is intertwined, leading to the submergence of crucial spatial semantic information, such as the intrinsic motion of key semantic objects and interactions among semantic objects. To address these limitations, it is imperative to explore alternative approaches. With the remarkable performance of general semantic segmentation models in visual representation, we introduce the general segmentation model SEEM into the video embedding paradigm, constructing a semantically structured representation from perceptual semantics to cognitive semantics. To more effectively utilize SEEM for structured video representation, we designed the Semantic Adapter (Sem-Adapter) as a bridge to connect the two models. Firstly, we design a Self-Motion Module (SMM) to pay attention to the self-motion of key semantic regions. Secondly, we propose a Mutual Relation Module (MRM) to construct the interactions between semantic regions. Extensive experiments on ActivityNet-1.3, THUMOS-14 and EPIC-Kitchens-100 reveal that our method significantly outperforms state-of-the-art methods under the same input modality, and our method improves the average mAP from 60.6% to 64.2% on THUMOS-14 with the same backbone. The code is available on https://***/shouxiaozixuan/semtad. © 1991-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Learning 3D Scene Analogies with Neural Contextual Scene Maps

arXiv

引用

arXiv 2025年

作者： Kim, Junho Bae, Gwangtak Lee, Eun Sun Kim, Young Min Dept. of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence and INMC Seoul National University Korea Republic of

Understanding scene contexts is crucial for machines to perform tasks and adapt prior knowledge in unseen or noisy 3D environments. As data-driven learning is intractable to comprehensively encapsulate diverse ranges of layouts and open spaces, we propose teaching machines to identify relational commonalities in 3D spaces. Instead of focusing on point-wise or object-wise representations, we introduce 3D scene analogies, which are smooth maps between 3D scene regions that align spatial relationships. Unlike well-studied single instance-level maps, these scene-level maps smoothly link large scene regions, potentially enabling unique applications in trajectory transfer in AR/VR, long demonstration transfer for imitation learning, and context-aware object rearrangement. To find 3D scene analogies, we propose neural contextual scene maps, which extract descriptor fields summarizing semantic and geometric contexts, and holistically align them in a coarse-to-fine manner for map estimation. This approach reduces reliance on individual feature points, making it robust to input noise or shape variations. Experiments demonstrate the effectiveness of our approach in identifying scene analogies and transferring trajectories or object placements in diverse indoor scenes, indicating its potential for robotics and AR/VR applications. © 2025, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

An Improved Vision Transformer Architecture-Based computer Recognition Platform for Lilium Classification

An Improved Vision Transformer Architecture-Based Computer R...

引用

IEEE International Conference on Consumer Electronics (ICCE)

作者： Xiang-Rui Huang Huai-Jhan Ke Szu-Yu Kuo Liang-Bi Chen Department of Electrical Engineering Master Program in Electrical Engineering and Computer Science National Penghu University of Science and Technology Penghu Taiwan Department of Computer Science and Information Engineering National Penghu University of Science and Technology Penghu Taiwan Department of Shipping and Transportation Management National Kaohsiung University of Science and Technology Kaohsiung Taiwan

ISBN: (数字)9798331521165

ISBN: (纸本)9798331521172

Lilies are popular in the global flower market, but consumers often lack information about specific varieties. To address this issue, this paper proposes a computer recognition platform based on the Vision Transformer (ViT) architecture. The proposed platform uses an improved vision transformer (ViT) architecture to classify different types of lilies, allowing consumers to access information and names of various Lilium species. The experimental results show that the proposed lily classification model achieved a 96.4% accuracy rate in classifying six lily species.

关键词： computer vision Electric potential Accuracy Computational modeling computer architecture computer applications Flowering plants Network architecture Transformers Consumer electronics

来源：评论

学校读者我要写书评

暂无评论

Fate-tox: fragment attention transformer for E(3)-equivariant multi-organ toxicity prediction

引用

Journal of Cheminformatics 2025年第1期17卷 1-16页

作者： Ha, Sumin Bang, Dongmin Kim, Sun Interdisciplinary Program in Artificial Intelligence Seoul National University Seoul Republic of Korea Interdisciplinary Program in Bioinformatics Seoul National University Seoul Republic of Korea AIGENDRUG Co. Ltd. Seoul Republic of Korea Department of Computer Science and Engineering Seoul National University Seoul Republic of Korea

Toxicity is a critical hurdle in drug development, often causing the late-stage failure of promising compounds. Existing computational prediction models often focus on single-organ toxicity. However, avoiding toxicity of an organ, such as reducing gastrointestinal side effects, may inadvertently lead to toxicity in another organ, as seen in the real case of rofecoxib, which was withdrawn due to increased cardiovascular risks. Thus, simultaneous prediction of multi-organ toxicity is a desirable but challenging task. The main challenges are (1) the variability of substructures that contribute to toxicity of different organs, (2) insufficient power of molecular representations in diverse perspectives, and (3) explainability of prediction results especially in terms of substructures or potential toxicophores. To address these challenges with multiple strategies, we developed FATE-Tox, a novel multi-view deep learning framework for multi-organ toxicity prediction. For variability of substructures, we used three fragmentation methods such as BRICS, Bemis-Murcko scaffolds, and RDKit Functional Groups to formulate fragment-level graphs so that diverse substructures can be used to identify toxicity for different organs. For insufficient power of molecular representations, we used molecular representations in both 2D and 3D perspectives. For explainability, our fragment attention transformer identifies potential 3D toxicophores using attention coefficients. Scientific contribution: Our framework achieved significant improvements in prediction performance, with up to 3.01% gains over prior baseline methods on toxicity benchmark datasets from MoleculeNet (BBBP, SIDER, ClinTox) and TDC (DILI, Skin Reaction, Carcinogens, and hERG), while the multi-task learning approach further enhanced performance by up to 1.44% compared to the single-task learning framework that had already surpassed these baselines. Additionally, attention visualization aligning with literature contributes to

关键词： Graph neural network Multi-task learning Toxicity prediction Transformer

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：