检索结果-南通市图书馆

Computer Modeling in Engineering & Sciences 2024年第9期140卷 2425-2448页

作者： md.Atiqur Rahman Mustavi Ibne Masum Khan md Hasib M.f.Mridha Sultan Alf.rhood Mejdl Saf.an Dunren Che Department of.Computer Science and Engineering Ahsanullah University of Science and TechnologyDhaka1208Bangladesh Department of.Computer Science and Sof.ware Engineering The University of Western AustraliaPerthWA 6009Australia Department of.Computer Science American International University-BangladeshDhaka1229Bangladesh Department of.Computer Science College of Computer and Information SciencesKing Saud UniversityP.O.Box 51178Riyadh11543Saudi Arabia School of.Computing Southern Illinois UniversityCarbondale62901USA

Brain tumors pose a signif.cant threat to human lives and have gained increasing attention as the tenth leading cause of.global *** study addresses the pressing issue of.brain tumor classif.cation using Magnetic resonance imaging(MRI).It f.cuses on distinguishing between Low-Grade Gliomas(LGG)and High-Grade Gliomas(HGG).LGGs are benign and typically manageable with surgical resection,while HGGs are malignant and more *** research introduces an innovative custom convolutional neural network(CNN)model,*** stands out as a lightweight CNN model compared to its *** research utilized the BraTS 2020 dataset f.r its *** with the gradient-boosting algorithm,GliomaCNN has achieved an impressive accuracy of.99.1569%.The model’s interpretability is ensured through SHapley Additive exPlanations(SHAP)and Gradient-weighted Class Activation Mapping(Grad-CAM++).They provide insights into critical decision-making regions f.r classif.cation *** challenges in identif.ing tumors in images without visible signs,the model demonstrates remarkable perf.rmance in this critical medical application,of.ering a promising tool f.r accurate brain tumor diagnosis which paves the way f.r enhanced early detection and treatment of.brain tumors.

关键词： Deep learning magnetic resonance imaging convolutional neural networks explainable AI boosting algorithm ablation

An Enhanced Hybrid Model Based on CNN and BiLSTM f.r Identif.ing Individuals via Handwriting Analysis

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2024年第8期140卷 1689-1710页

作者： md.Abdur Rahim f.hmid Al f.rid Abu Saleh Musa Miah Arpa Kar Puza md.Nur Alam md.Najmul Hossain Sarina Mansor Hezerul Abdul Karim Department of.Computer Science and Engineering Pabna University of Science and TechnologyPabna6600Bangladesh f.culty of.Engineering Multimedia UniversityCyberjaya63100Malaysia Department of.Computer Science and Engineering Bangladesh Army University of Science and TechnologySaidpur5311Bangladesh Department of.Mathematics Pabna University of Science and TechnologyPabna6600Bangladesh Department of.Electrical Electronic and Communication EngineeringPabnaUniversity of Science and TechnologyPabna6600Bangladesh Department of.Electrical and Communication Engineering Pabna University of Science and TechnologyPabna6600Bangladesh

Handwriting is a unique and signif.cant human f.ature that distinguishes them f.om one *** are many researchers have endeavored to develop writing recognition systems utilizing specif.c signatures or symbols f.r person identif.cation through ***,such systems are susceptible to f.rgery,posing security *** response to these challenges,we propose an innovative hybrid technique f.r individual identif.cation based on independent handwriting,eliminating the reliance on specif.c signatures or *** response to these challenges,we propose an innovative hybrid technique f.r individual identif.cation based on independent handwriting,eliminating the reliance on specif.c signatures or *** innovative method is intricately designed,encompassing f.ve distinct phases:data collection,preprocessing,f.ature extraction,signif.cant f.ature selection,and *** key advancement lies in the creation of.a novel dataset specif.cally tailored f.r Bengali handwriting(BHW),setting the f.undation f.r our comprehensive ***-preprocessing,we embarked on an exhaustive f.ature extraction process,encompassing integration with kinematic,statistical,spatial,and composite *** meticulous amalgamation resulted in a robust set of.91 *** enhance the ef.iciency of.our system,we employed an analysis of.variance(ANOVA)f.test and mutual inf.rmation scores approach,meticulously selecting the most pertinent *** the identif.cation phase,we harnessed the power of.cutting-edge deep learning models,notably the Convolutional Neural Network(CNN)and Bidirectional Long Short-Term Memory(BiLSTM).These models underwent rigorous training and testing to accurately discern individuals based on their handwriting ***,our methodology introduces a groundbreaking hybrid model that synergizes CNN and BiLSTM,capitalizing on f.ne motor f.atures f.r enhanced individual ***,our experimental results unde

关键词： Bengali handwriting(BHW) person identif.cation convolutional neural network(CNN) BiLSTM

Probiotics ameliorate chromium-induced growth retardation and stress in Indian major carp rohu, Labeo rohita

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Emerging Contaminants 2024年第2期10卷 57-64页

作者： Zannatul f.rdous SMRaf.quzzaman md Shahjahan Laboratory of.f.sh Ecophysiology Department of Fisheries ManagementBangladesh Agricultural UniversityMymensingh2202Bangladesh Department of.f.sheries Biology&Aquatic Environment Bangabandhu Sheikh Mujibur Rahman Agricultural UniversityGazipur1706Bangladesh

Chromium(Cr)is one of.the most prominent contaminants due to its cumulative negative ef.ects on *** may be able to mitigate the unf.vorable ef.ects of.Cr in *** this study,the rohu Labeo rohita(3.02±0.37 g)were exposed to Cr with or without probiotics,representing control(C;without Cr or probiotics),treatment one(T1;Cr at a dose of.4 mg/L),treatment two(T2;probiotics at a dose of.1 ml/L)and treatment three(T3;Cr at a dose of.4 mg/L+probiotics at a dose of.1 ml/L)f.r 42 *** results showed that weight gain(WG)and specif.c growth rate(SGR)were inhibited signif.cantly in T1,while those substantially increased in T2(probiotics)compared to *** the other hand,WG and SGR were regained as compared to C in *** signif.cantly lowest f.ed conversion ratio(f.R)was f.und in T2,and moderate f.R was recorded in C and T3,while it was signif.cantly highest in *** mortality was observed in C and T2,while 20%mortality was recorded in T1,and decreased mortality was recorded in T3(10%).Exposure to Cr(T1)caused lower hemoglobin and higher glucose levels;however,the use of.probiotics reversed these ef.ects in T3(Cr+probiotics).The f.equencies of.nuclear(notched nuclei,nuclear bridges,and karyopyknosis)and cellular(tear drop shaped,f.sion,and twin)abnormalities of.erythrocytes increased in the Cr-exposed group(T1);however,these were recovered af.er probiotic administration(T3).The f.ndings of.the present study suggest that probiotics play a benef.cial role in reducing the toxicities of.Cr in rohu.

关键词： Contaminants Pollution Stress Erythrocytes Hemoglobin Probiotics

Multi-species probiotics enhance survival,growth,intestinal microbiota and disease resistance of.rohu(Labeo rohita)larvae

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Water Biology and Security 2024年第1期3卷 71-79页

作者： Zannatul f.rdous md Kabir Hossain md Hadiuzzaman SMRaf.quzzaman KM Abdul Halim Tanvir Rahman md Ali Reza f.ruk Zulhisyam Abdul Kari md Shahjahan Laboratory of.f.sh Ecophysiology Department of Fisheries ManagementBangladesh Agricultural UniversityMymensingh2202Bangladesh Department of.Aquaculture Bangladesh Agricultural UniversityMymensingh2202Bangladesh Department of.f.sheries Matshoy BhabanDhakaBangladesh Department of.f.sheries Biology&Aquatic Environment Bangabandhu Sheikh Mujibur Rahman Agricultural UniversityGazipur1706Bangladesh Department of.Agricultural Sciences Faculty of Agro-Based IndustryUniversiti Malaysia KelantanJeli Campus17600JeliKelantanMalaysia

Multi-strain probiotics provide the most dependable approaches to improve health,immune response,and disease resistance in f.rmed *** the present study,we examined the ef.ects of.multi-species probiotics on the survival,growth,immune response,and disease resistance of.rohu(Labeo rohita)*** hatched larvae f.om the day of.f.rst f.eding(average weight of.0.003 g)were reared with multi-species probiotics having a combination of.Bacillus subtilis(109 colony f.rming units(cf.)/mL)and Lactobacillus spp.(Lactobacillus plantarum,Lactobacillus buchneri-1011 cf./mL)in water containing doses of.0(control-C),0.5 mL/L(treatment 1-T1),and 1.0 mL/L(treatment 2-T2)in triplicates f.r 90 *** the experiment,a challenge test was perf.rmed to assess the f.sh's resistance to pathogenic Aeromonas *** higher survival was recorded in larvae of.treated groups(87%in T2 and 79%in T1)compared to the control(62%).Signif.cantly higher growth perf.rmance(weight gain and specif.c growth rate—SGR)was shown by the probiotic-treated larval groups compared to the *** supplementation resulted in signif.cantly higher counts of.total viable colony(TVC)and lactic acid bacteria(LAB)in the *** immunological parameters(mucosal f.ld f.ttening,goblet cell abundance,expansion of.lamina propria and enterocytes)of.the gut were signif.cantly better in probiotic-treated *** liver of.treated f.sh showed irregular shape nuclei turning into regular shape and reducing spaces between the hepatic ***-treated f.sh had the highest post-challenge survival rate(90%)against *** *** erythrocytes of.challenged f.sh treated with probiotics had signif.cantly lower f.equencies of.various nuclear and cellular *** f.ndings suggest that multi-species probiotic supplements could improve the survival,growth,health status,and immune response of.rohu in the early stages of.its development.

关键词： Aquaculture Disease resistance Immunity Labeo rohita Probiotics

Ef.ectiveness of.hybrid ensemble machine learning models f.r landslide susceptibility analysis:Evidence f.om Shimla district of.North-west Indian Himalayan region

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of.Mountain Science 2024年第7期21卷 2368-2393页

作者： SHARMA Aastha SAJJAD Haroon RAHAMAN md Hibjur SAHA Tamal Kanti BHUYAN Nirsobha Department of.Geography Faculty of SciencesJamia Millia IslamiaNew Delhi-110025India

The Indian Himalayan region is f.equently experiencing climate change-induced ***,landslide susceptibility assessment assumes greater signif.cance f.r lessening the impact of.a landslide *** paper makes an attempt to assess landslide susceptibility in Shimla district of.the northwest Indian Himalayan *** examined the ef.ectiveness of.random f.rest(Rf.,multilayer perceptron(MLP),sequential minimal optimization regression(SMOreg)and bagging ensemble(B-Rf.BSMOreg,B-MLP)models.A landslide inventory map comprising 1052 locations of.past landslide occurrences was classif.ed into training(70%)and testing(30%)*** site-specif.c inf.uencing f.ctors were selected by employing a multicollinearity *** relationship between past landslide occurrences and inf.uencing f.ctors was established using the f.equency ratio *** ef.ectiveness of.machine learning models was verif.ed through perf.rmance *** landslide susceptibility maps were validated by the area under the receiver operating characteristic curves(ROC-AUC),accuracy,precision,recall and *** key perf.rmance metrics and map validation demonstrated that the BRf.model(correlation coef.icient:0.988,mean absolute error:0.010,root mean square error:0.058,relative absolute error:2.964,ROC-AUC:0.947,accuracy:0.778,precision:0.819,recall:0.917 and f.1 score:0.865)outperf.rmed the single classif.ers and other bagging ensemble models f.r landslide *** results show that the largest area was f.und under the very high susceptibility zone(33.87%),f.llowed by the low(27.30%),high(20.68%)and moderate(18.16%)susceptibility *** f.ctors,namely average annual rainf.ll,slope,lithology,soil texture and earthquake magnitude have been identif.ed as the inf.uencing f.ctors f.r very high landslide *** texture,lineament density and elevation have been attributed to high and moderate ***,the study calls f.r devising suitable landslide mitig

关键词： Landslide susceptibility Site-specif.c f.ctors Machine learning models Hybrid ensemble learning Geospatial techniques Himalayan region

Enhancing Internet of.Things Intrusion Detection Using Artif.cial Intelligence

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computers, Materials and Continua 2024年第1期81卷 1-23页

作者： Bar, Shachar Prasad, P.W.C. Sayeed, md Shohel School of.Computing Mathematics Charles Sturt University Bathurst NSW 2795 Australia International School Duy Tan University Da Nang 550000 Viet Nam f.culty of.Inf.rmation Science and Technology Multimedia University Melaka 75450 Malaysia

Escalating cyber security threats and the increased use of.Internet of.Things (IoT) devices require utilisation of.the latest technologies available to supply adequate protection. The aim of.Intrusion Detection Systems (IDS) is to prevent malicious attacks that corrupt operations and interrupt data f.ow, which might have signif.cant impact on critical industries and inf.astructure. This research examines existing IDS, based on Artif.cial Intelligence (AI) f.r IoT devices, methods, and techniques. The contribution of.this study consists of.identif.cation of.the most ef.ective IDS systems in terms of.accuracy, precision, recall and f.-score; this research also considers training time. Results demonstrate that Graph Neural Networks (GNN) have several benef.ts over other traditional AI f.ameworks through their ability to achieve in excess of.99% accuracy in a relatively short training time, while also capable of.learning f.om network traf.ic the inherent characteristics of.dif.erent cyber-attacks. These f.ndings identif. the GNN (a Deep Learning AI method) as the most ef.icient IDS system. The novelty of.this research lies also in the linking between high yielding AI-based IDS algorithms and the AI-based learning approach f.r data privacy protection. This research recommends f.derated Learning (f.) as the AI training model, which increases data privacy protection and reduces network data f.ow, resulting in a more secure and ef.icient IDS solution. © 2024 The Authors. Published by Tech Science Press.

关键词： Network intrusion

Improving Prediction Ef.iciency of.Machine Learning Models f.r Cardiovascular Disease in IoST-Based Systems through Hyperparameter Optimization

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第9期80卷 3485-3506页

作者： Tajim md.Niamat Ullah Akhund Waleed M.Al-Nuwaiser Department of.Computer Science and Engineering(CSE) Daffodil International UniversityDhaka1216Bangladesh Graduate School of.Science and Engineering Saga UniversitySaga8408502Japan Computer Science Department College of Computer and Information SciencesImam Mohammad Ibn Saud Islamic University(IMSIU)Riyadh11623Saudi Arabia

This study explores the impact of.hyperparameter optimization on machine learning models f.r predicting cardiovascular disease using data f.om an IoST(Internet of.Sensing Things)*** distinct machine learning approaches were implemented and systematically evaluated bef.re and af.er hyperparameter *** improvements were observed across various models,with SVM and Neural Networks consistently showing enhanced perf.rmance metrics such as f.-Score,recall,and *** study underscores the critical role of.tailored hyperparameter tuning in optimizing these models,revealing diverse outcomes among *** Trees and Random f.rests exhibited stable perf.rmance throughout the *** enhancing accuracy,hyperparameter optimization also led to increased execution *** representations and comprehensive results support the f.ndings,conf.rming the hypothesis that optimizing parameters can ef.ectively enhance predictive capabilities in cardiovascular *** research contributes to advancing the understanding and application of.machine learning in healthcare,particularly in improving predictive accuracy f.r cardiovascular disease management and intervention strategies.

关键词： Internet of.sensing things(IoST) machine learning hyperparameter optimization cardiovascular disease prediction execution time analysis perf.rmance analysis wilcoxon signed-rank test

A Comprehensive Survey f.r Privacy-Preserving Biometrics: Recent Approaches, Challenges, and f.ture Directions

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第2期78卷 2087-2110页

作者： Shahriar md Arman Tao Yang Shahadat Shahed Alanoud AlMazroa Af.aa Attiah Linda Mohaisen School of.Computer Science and Technology Zhejiang Gongshang UniversityHangzhou310018China School of.Computer Science and Technology Hangzhou Dianzi UniversityHangzhou310018China College of.Computer and Inf.rmation Sciences Princess Nourah bint Abdulrahman UniversityRiyadh84428Saudi Arabia f.culty of.Computing and Inf.rmation Technology King Abdulaziz UniversityJeddah21589Saudi Arabia

The rapid growth of.smart technologies and services has intensif.ed the challenges surrounding identity authenti-cation *** credentials are increasingly being used f.r verif.cation due to their advantages over traditional methods,making it crucial to saf.guard the privacy of.people’s biometric data in various *** paper of.ers an in-depth exploration f.r privacy-preserving techniques and potential threats to biometric *** proposes a noble and thorough taxonomy survey f.r privacy-preserving techniques,as well as a systematic f.amework f.r categorizing the f.eld’s existing *** review the state-of.the-art methods and address their advantages and limitations in the context of.various biometric modalities,such as f.ce,f.ngerprint,and eye *** survey encompasses various categories of.privacy-preserving mechanisms and examines the trade-of.s between security,privacy,and recognition perf.rmance,as well as the issues and f.ture research *** aims to provide researchers,prof.ssionals,and decision-makers with a thorough understanding of.the existing privacy-preserving solutions in biometric recognition systems and serves as the f.undation of.the development of.more secure and privacy-preserving biometric technologies.

关键词： Biometric modalities biometric recognition biometric security privacy-preserving security threats

Computer vision-based six layered ConvNeural network to recognize sign language f.r both numeral and alphabet signs

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Biomimetic Intelligence & Robotics 2024年第1期4卷 45-58页

作者： Muhammad Aminur Rahaman Kabiratun Ummi Oyshe Prothoma Khan Chowdhury Tanoy Debnath Anichur Rahman md.Saikat Islam Khan Department of.Computer Science and Engineering Green University of BangladeshDhakaBangladesh Department of.Computer Science and Engineering National Institute of Textile Engineering and Research(NITER)Constituent Institute of Dhaka UniversityDhakaBangladesh Department of.Computer Science and Engineering Mawlana Bhashani Science and Technology UniversityTangailBangladesh

People who have trouble communicating verbally are of.en dependent on sign language,which can be dif.icult f.r most people to understand,making interaction with them a dif.icult *** Sign Language Recognition(SLR)system takes an input expression f.om a hearing or speaking-impaired person and outputs it in the f.rm of.text or voice to a normal *** existing study related to the Sign Language Recognition system has some drawbacks,such as a lack of.large datasets and datasets with a range of.backgrounds,skin tones,and *** research ef.iciently f.cuses on Sign Language Recognition to overcome previous *** importantly,we use our proposed Convolutional Neural Network(CNN)model,“ConvNeural”,in order to train our ***,we develop our own datasets,“BdSL_OPSA22_STATIC1”and“BdSL_OPSA22_STATIC2”,both of.which have ambiguous backgrounds.“BdSL_OPSA22_STATIC1”and“BdSL_OPSA22_STATIC2”both include images of.Bangla characters and numerals,a total of.24,615 and 8437 images,***“ConvNeural”model outperf.rms the pre-trained models with accuracy of.98.38%f.r“BdSL_OPSA22_STATIC1”and 92.78%f.r“BdSL_OPSA22_STATIC2”.f.r“BdSL_OPSA22_STATIC1”dataset,we get precision,recall,f.-score,sensitivity and specif.city of.96%,95%,95%,99.31%,and 95.78%***,in case of.BdSL_OPSA22_STATIC2”dataset,we achieve precision,recall,f.-score,sensitivity and specif.city of.90%,88%,88%,100%,and 100%respectively.

关键词： Conv NeuralSign language CNN Static f.ature extraction Convolution2D f.lly connected layer Dropout