检索结果-内蒙古大学图书馆

dIGITAL dISCOVErY 2024年第3期3卷 558-572页

作者： Torren-Peraire, Paula Hassen, Alan Kai Genheden, Samuel Verhoeven, Jonas Clevert, djork-Arne Preuss, Mike Tetko, Igor V. Helmholtz Munich Deutsch Forschungszentrum Julich Inst Struct Biol Mol Targets & Therapeut Ctr Neuherberg Germany In Silico Discovery Janssen Res & Dev Janssen Pharmaceut NV Beerse Belgium Leiden Univ Leiden Inst Adv Comp Sci Leiden Netherlands Machine Learning Res Pfizer Res & Dev Berlin Germany Mol AI Discovery Sci R&D AstraZeneca Gothenburg Sweden

retrosynthesis consists of breaking down a chemical compound recursively step-by-step into molecular precursors until a set of commercially available molecules is found with the goal to provide a synthesis route. Its two primary research directions, single-step retrosynthesis prediction, which models the chemical reaction logic, and multi-step synthesis planning, which tries to find the correct sequence of reactions, are inherently intertwined. Still, this connection is not reflected in contemporary research. In this work, we combine these two major research directions by applying multiple single-step retrosynthesis models within multi-step synthesis planning and analyzing their impact using public and proprietary reaction data. We find a disconnection between high single-step performance and potential route-finding success, suggesting that single-step models must be evaluated within synthesis planning in the future. Furthermore, we show that the commonly used single-step retrosynthesis benchmark dataset USPTO-50k is insufficient as this evaluation task does not represent model scalability or performance on larger and more diverse datasets. For multi-step synthesis planning, we show that the choice of the single-step model can improve the overall success rate of synthesis planning by up to +28% compared to the commonly used baseline model. Finally, we show that each single-step model finds unique synthesis routes, and differs in aspects such as route-finding success, the number of found synthesis routes, and chemical validity. Synthesis planning relies on retrosynthesis models, yet this relationship is under-analyzed. We investigate the effect of contemporary single-step models trained on public and proprietary reaction data to analyze the synthesis routes produced.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhanced Brain Tumor diagnosis through Multimodal Fusion and 3d Convolutional Neural Networks 2

Enhanced Brain Tumor Diagnosis through Multimodal Fusion and...

引用

2nd International Conference on Smart Technologies for Smart Nation, SmartTechCon 2023

作者： Belinda, Carmel Mary M.J. Alex david, S. Begum, Almas Kannan, E. Thangarasu, Gunasekar Department of Artificial Intelligence and Machine learning Saveetha School of Engineering Chennai India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai India Department of Professional Industry Driven Education MAHSA University Saujana Putra Malaysia

ISBN: (纸本)9798350305418

Brain tumor detection has enhanced using a Multi-CNNs with Multimodal Information Fusion method has been proposed in this work. The proposed method leverages multiple modalities, like T1-weighted, T2-weighted, and FLAIr images, to create a fused representation of the input data. A deep neural network architecture using 3d Convolutional Neural Networks (3d-CNNs) is constructed to capture spatial and volumetric information present in medical images. The method also incorporates a real normalization layer and a weighted loss function to improve the learning process and prioritize accurate detection and localization of tumor lesions. Evaluating our method on a Kaggle dataset of 253 images (98 without tumor, 155 with tumor), we achieve superior results compared to existing approaches. Our proposed method attains an accuracy of 98.9%, sensitivity of 99.2%, specificity of 99.1%, recall of 98.7%, precision of 99.3%, and F1 score of 98.9%. Leveraging multimodal fusion, 3d-CNNs, and additional enhancements, our study contributes to more reliable brain tumor diagnosis, benefiting patients and healthcare providers. © 2023 IEEE.

关键词： 3d Convolutional Neural Networks (3d-CNNs) Accuracy enhancement Multi-CNNs Multimodal Tumor detection

来源：评论

学校读者我要写书评

暂无评论

Question Answering FAQ Bot Using Natural Language Processing 11th

Question Answering FAQ Bot Using Natural Language Processing

引用

11th International Conference on Intelligent Computing and Applications, ICrTC 2023

作者： Vijay, K. Sowmia, K.r. ravikumar, S. Arockia raj, Y. Department of Computer Science and Engineering Rajalakshmi Engineering College Chennai India Department of Artificial Intelligence and Machine Learning Rajalakshmi Engineering College Chennai India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai India Department of Computer Science and Engineering PSNA College of Engineering and Technology Dindigul India

ISBN: (纸本)9789819717231

The BOT retorts with the precise results that help the user for the query and also requests them to rate the response to provide better results. The [frequently answered questions] BOT is a program developed through machine learning and natural language processing. It classifies the queries and results as key-value pairs and maps the response to the user query. It extricates itself by providing additional capabilities such as NLU processes and reinforcement learning. NLU helps the bot to make it convenient for an array of languages to meet the diverse expectations. reinforcement learning retrieves user rating to provide better performance and search results. The project was developed due to a concern about fetching meticulous responses for specific service-related queries. The overall satisfaction rate of users in getting responses to the questions they post is devastating. The root cause for the proposed project has aimed at providing an environment for a people to find their answers at better accuracy and also respond to their satisfaction. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Toward enhancement of antibody thermostability and affinity by computational design in the absence of antigen

引用

MABS 2024年第1期16卷 2362775页

作者： Hutchinson, Mark ruffolo, Jeffrey A. Haskins, Nantaporn Iannotti, Michael Vozza, Giuliana Pham, Tony Mehzabeen, Nurjahan Shandilya, Harini rickert, Keith Croasdale-Wood, rebecca damschroder, Melissa Fu, Ying dippel, Andrew Gray, Jeffrey J. Kaplan, Gilad AstraZeneca Biol Engn R&D Gaithersburg MD 20852 USA Johns Hopkins Univ Program Mol Biophys Baltimore MD 21218 USA Profluent Bio Machine Learning Berkeley CA USA Amgen Inc Prot Engn R&D Rockville MD USA Honigman LLP Intellectual Property Washington DC USA AstraZeneca Biopharmaceut Dev R&D Cambridge England AstraZeneca Biol Engn R&D Cambridge England Johns Hopkins Univ Dept Chem & Biomol Engn Baltimore MD 21218 USA

Over the past two decades, therapeutic antibodies have emerged as a rapidly expanding domain within the field of biologics. In silico tools that can streamline the process of antibody discovery and optimization are critical to support a pipeline that is growing more numerous and complex every year. High-quality structural information remains critical for the antibody optimization process, but antibody-antigen complex structures are often unavailable and in silico antibody docking methods are still unreliable. In this study, deepAb, a deep learning model for predicting antibody Fv structure directly from sequence, was used in conjunction with single-point experimental deep mutational scanning (dMS) enrichment data to design 200 potentially optimized variants of an anti-hen egg lysozyme (HEL) antibody. We sought to determine whether deepAb-designed variants containing combinations of beneficial mutations from the dMS exhibit enhanced thermostability and whether this optimization affected their developability profile. The 200 variants were produced through a robust high-throughput method and tested for thermal and colloidal stability (T-onset, T-m, T-agg), affinity (K-d) relative to the parental antibody, and for developability parameters (nonspecific binding, aggregation propensity, self-association). Of the designed clones, 91% and 94% exhibited increased thermal and colloidal stability and affinity, respectively. Of these, 10% showed a significantly increased affinity for HEL (5- to 21-fold increase) and thermostability (>2.5C increase in T-m1), with most clones retaining the favorable developability profile of the parental antibody. Additional in silico tests suggest that these methods would enrich for binding affinity even without first collecting experimental dMS measurements. These data open the possibility of in silico antibody optimization without the need to predict the antibody-antigen interface, which is notoriously difficult in the absence of crystal struc

关键词： Antibodies computational design developability machine learning thermostability

来源：评论

学校读者我要写书评

暂无评论

Structural and micromagnetic modeling of the magnetic binder phase in WC-Co cemented carbides

Structural and micromagnetic modeling of the magnetic binder...

引用

2023 IEEE International Magnetic Conference - Short Papers, INTErMAG Short Papers 2023

作者： Breth, Leoni Fischbacher, Johann Kovacs, Alexander Oezelt, Harald Schrefl, Thomas Czettl, Christoph Kuehrer, Saskia Pachlhofer, Julia Schwarz, Maria Weirather, Thomas Brueckl, Hubert University for Continuing Education Krems Dept. Integrated Sensor Systems Wiener Neustadt Austria Christian Doppler Laboratory for Magnet Design Through Physics Informed Machine Learning University for Continuing Education Krems Wiener Neustadt Austria Ceratizit Austria GmbH R&d Cutting Tools Reutte Austria

ISBN: (纸本)9798350338362

The magnetic properties of tungsten carbide with cobalt as a binder have long been used in industry to characterize the microstructure of this hard metal. We present bottom-up structural and micromagnetic modeling of WC-Co, from which we obtain coercive fields that compare well to experimental data. Future work will use experimentally obtained microstructure information as input to the structural modeling. © 2023 IEEE.

关键词： Microstructure

来源：评论

学校读者我要写书评

暂无评论

Secured Mobile Attendance Management System Integrating Multi-Layered Authentication

Secured Mobile Attendance Management System Integrating Mult...

引用

International Conference on Trends in Electronics and Informatics (ICEI)

作者： P. Pandiaraja Muthu Kumaran V Charles Prabu V Iraniya Pandiyan M S. Vanarasan C. Selvarathi Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Institute of Science and Technology Chennai Department of Artificial Intelligence and Machine Learning Sri Sairam College of Engineering Banglore

ISBN: (数字)9798331544607

ISBN: (纸本)9798331544614

Ensuring precise and valid attendance tracking has gained utmost prominence in educational institutions and corporate organizations. The traditional attendance tracking system is easily susceptible to manipulation and fraud, thus compromising the integrity of the data. In this paper, we proposed a secured system that integrates multilayered authentication with feedback module to ensure reliability, accuracy, and a secure attendance marking system. The first layer MAC address verification, which authenticates the registered mobile device to prevent unauthorized access from other devices, the second layer Face recognition using Face Net, ensuring that only the legitimate user can mark attendance, and the third layer Geolocation verification, which verifies the user's physical presence at the designated location by matching real-time GPS coordinates. Users can report issues and concerns that occurred in any of these verification layers, in the feedback module, later an administrator reviews and validates the reports before final attendance approval. This proposed solution simplifies the attendance management procedure, improves security, and reduces attendance fraud. Which is particularly beneficial for educational institutions, corporate organizations, and other sectors requiring efficient attendance tracking system.

关键词： Accuracy Face recognition Geology Authentication Organizations real-time systems Mobile handsets Fraud reliability Security

来源：评论

学校读者我要写书评

暂无评论

Smart solar power Conversion: Leveraging deep learning MPPT and hybrid cascaded h-bridge multilevel inverters for optimal efficiency

引用

BIOMEdICAL SIGNAL PrOCESSING ANd CONTrOL 2025年 105卷

作者： ramu, K. Sreenivasulu, Gopu dixit, rinku Sharma Choudhary, Shailee Lohmor rao, Katakam Venkateswara Suman, Sanjay Kumar Shuaib, Mohammed rajaram, A. Vinayaka Mission Res Fdn VMRF Deemed Univ Aarupadai Veedu Inst Technol AVIT Dept CSE OMR Rd Chennai Tamil Nadu India Rajeev Gandhi Mem Coll Engn & Technol Autonomous Dept Civil Engn Nandyal 518501 Andhra Pradesh India New Delhi Inst Management Artificial Intelligence & Machine Learning Delhi India New Delhi Inst Management Dept Artificial Intelligence & Machine Learning Delhi India Koneru Lakshmaiah Educ Fdn Dept Comp Sci & Engn Vaddeswaram Guntur AP India St Martins Engn Coll Secunderabad Dept ECE Secunderabad Telangana India St Martins Engn Coll R&D Secunderabad Telangana India Jazan Univ Coll Engn & Comp Sci Dept Comp Sci Jazan Saudi Arabia EGS Pillay Engn Coll Dept Elect & Commun Engn Nagapattinam 611002 Tamil Nadu India

In the era of smart energy systems, maximizing the efficiency of solar power conversion is more crucial than ever. This research presents a new solar power conversion system that utilizes advanced deep learning maximum power point tracking integrated with a novel Hybrid Cascaded H-Bridge Multilevel Inverter to provide optimal energy extraction and quality AC power. Through using of the deep Neural Network, the system effectively follows operation at the maximum power point of the photovoltaic assembly to different stochastic characteristics like solar irradiation and temperature to improve power harvesting. Inverter is available in 5-level, 9-level, and 17-level configurations to maintain systems simplicity and yet have low Total Harmonic distortion. Adopting Space Vector Pulse Width Modulation, the inverter produces power that is AC and the respective Total Harmonic distortion values are 3.1 %, 2.5 %, and 1.6 % for the 5-level, 9-level, and 17-level configurations, respectively. The proposed deep learning-based maximum power point tracking algorithm increases the system efficiency by approximately 15 % higher than conventional techniques such as P&O. Claims on both replication and experiments show that the possible peak system efficiency is as high as 98.6 %, while in the maximum power point tracking, averaging time converges in 35 ms. The presented integrated system is an efficient, modular concept for residential as commercial scale PV systems, maximizes the extracted power, particularly in rural locations, and ensures dependable power for medical equipment with premium quality AC with low total harmonic distortion.

关键词： Solar Power Conversion Hybrid Cascaded H-Bridge Multilevel Inverter deep learning Maximum Power Point Tracking Space Vector PWM Total Harmonic distortion Sustainability

来源：评论

学校读者我要写书评

暂无评论

A Hybrid deep learning Approach for Early detection of diabetic retinopathy 3

A Hybrid Deep Learning Approach for Early Detection of Diabe...

引用

3rd International Conference on Electronics and renewable Systems, ICEArS 2025

作者： Mary Vespa, M. Shajeena, J. Shiny, r.M. Gayathri devi, K. Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering Chennai India Srm Institute of Science and Technology Department of Computer Science and Engineering Tiruchirapalli Campus India St Joseph's College of Engineering Department of Computer Science and Engineering Chennai India Rmd Engineering College Department of Artificial Intelligence and Machine Learning Chennai India

ISBN: (数字)9798331509675

ISBN: (纸本)9798331509675

diabetes is a malnutrition that results from elevated glucose levels. diabetes eventually results in diabetic retinopathy (dr), a retinal condition that significantly impairs vision. It is an insulin-dependent difficulty that impacts ocular health. The receptive to cells at the back of the individual retina's veins gets injured, which is the cause of it. A favorable prognosis for this condition depends on its early identification. In this paper, the application of Convolutional Neural Network (CNN) with Long Short Term Memory (LSTM) on color fundus images is used for diabetic retinopathy is applied for understanding the task. The goal of the proposed dr identification method is to use deep learning to automatically identify the problem. An hybrid model that was trained and tested to distinguish between healthy and dr-affected fundi was able to achieve an accuracy of roughly 99.07% using fundus images from diabetic patients that were available in MESSIdOr. © 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

Smart Emergency Management System Using raspberry Pi Pico, GPS and GSM Modules 4

Smart Emergency Management System Using Raspberry Pi Pico, G...

引用

4th International Conference on Sentiment Analysis and deep learning, ICSAdL 2025

作者： Govinda rajulu, G. Sharmila, L. Venkatesan, d. Soumya, E. Prakash, Sargu Chandra Priyanka, Ms. P. Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Avadi Department of Computer Science and Engineering Tamil Nadu Chennai India St. Martin's Engineeringollege Dept. of Artificial Intelligence and Machine Learning Telangana Dhulapally Secunderabad India St. Martin's Engineeringollege Deptarment of Cse Telangana Dhulapally Secunderabad India Telangana Dhulapally Secunderabad India

ISBN: (纸本)9798331523923

This paper proposes a GSM-based children's rescue system designed for earthquake emergencies, emphasizing cost-effectiveness and simplicity while ensuring rapid child rescue. The system operates in three key stages: data acquisition, classification, and alert notification. Sensors collect temperature, pH, and heart rate data, which are then digitized and classified to assess the child's condition. If abnormal values are detected, an alert notification is sent via GSM to parents and rescue teams, providing real-time location data through GPS. The hardware implementation, based on raspberry Pi, converts analog sensor signals into digital data for processing. The system's efficacy is validated through various scenarios, demonstrating its ability to detect critical conditions and promptly trigger alerts, enhancing child safety during disasters. © 2025 IEEE.

关键词： Earthquakes

来源：评论

学校读者我要写书评

暂无评论

Insights into Target Gas-Oxygen Interactions in Highly Sensitive Gas Sensors Using data-driven Methods

引用

ACS SENSOrS 2024年第11期9卷 5848-5855页

作者： Kim, Kyusung Pornaroontham, Phuwadej Yun, Hojung Kim, Sungmin Choi, Pilgyu Masuda, Yoshitake Nagoya Univ Inst Mat Innovat Inst Innovat Future Soc Nagoya 4648601 Japan Chulalongkorn Univ Fac Sci Dept Chem Technol Bangkok 10330 Thailand Chulalongkorn Univ Machine learning Polymers & Mat Discovery Res Unit Bangkok 10330 Thailand Nagoya Univ Fac Engn Dept Sustainable Mat & Technol Ind Nagoya 4648601 Japan Korea Inst Ind Technol Surface R&D Grp 156 Gaetbeol Ro Incheon 21999 South Korea Natl Inst Adv Ind Sci & Technol 2266-98 AnagahoraMoriyama Nagoya 4638560 Japan

In the gas-sensing mechanism of a metal-oxide-semiconductor (n-type) gas sensor, oxygen adsorption or desorption on the oxide surface leads to an increase or decrease in the resistance of the gas sensor system. Additionally, oxygen can be adsorbed again at the location where initially adsorbed oxygen reacted with the target gas. Thus, the adsorption-desorption equilibrium of the reducing gas on the oxide surface is a significant factor in determining the sensitivity and reaction rate. In particular, for ultralow-concentration gas measurements, the relative concentration of oxygen was very high. To design an ultrasensitive gas sensor, not only the reaction of the target gas but also the competing reaction between the target gas and oxygen must be considered. Although qualitative investigations of these complex relationships have been performed according to the gas concentration and flow rate, reliable quantitative results are limited. In this study, a quantitative approach was used to understand the correlation between oxygen and a target gas by applying data analysis methods. We investigated the behavior of oxygen and the target molecules depending on the gas concentration and flow rate using the parts per billion level of the acetone gas sensor. Initial response data according to various detection conditions were processed using principal component analysis and K-means clustering;as a result, four types of reaction behaviors were inferred for 15 types of reaction conditions. Furthermore, the response time, depending on the detection conditions, can be distinguished using the suggested categorization. Our investigation suggests a possibility beyond simple optimization through the data analysis of the gas-sensing results.

关键词： SnO2 nanosheets gas sensors flow rate clustering

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：