10 research outputs found
Wavelet Attention VGG19 and XGBOOST for Classification of Skin Disease
No full text<p><strong>Abstract:</strong> This research paper introduces a novel framework for skin disease classification, combining Wavelet Attention VGG19 and XGBoost algorithms. Wavelet Attention VGG19 leverages the power of deep learning and wavelet attention mechanisms to extract discriminative features from skin lesion images, while XGBoost, a gradient boosting technique, complements the feature extraction capabilities with its ability to handle complex data relationships. The integration of these methodologies aims to improve accuracy and resilience in binary skin disease classification. The two goals of this study are to first improve feature learning and representation from skin lesion images by introducing wavelet attention into the VGG19 architecture, and second to improve classification performance by utilising XGBoost's ensemble and generalisation capabilities.</p>
<p><strong>Keywords:</strong> Wavelet Attention, Xgboost, deep learning, convolutional neural network, skin disease classification, VGG19.</p>
<p><strong>Title:</strong> Wavelet Attention VGG19 and XGBOOST for Classification of Skin Disease</p>
<p><strong>Author:</strong> Samaila Audu, Ali Ahmad Aminu</p>
<p><strong>International Journal of Computer Science and Information Technology Research</strong></p>
<p><strong>ISSN 2348-1196 (print), ISSN 2348-120X (online)</strong></p>
<p><strong>Vol. 11, Issue 4, October 2023 - December 2023</strong></p>
<p><strong>Page No: 5-13</strong></p>
<p><strong>Research Publish Journals</strong></p>
<p><strong>Website: www.researchpublish.com</strong></p>
<p><strong>Published Date: 07-October-2023</strong></p>
<p><strong>DOI: <a href="https://doi.org/10.5281/zenodo.8416714">https://doi.org/10.5281/zenodo.8416714</a></strong></p>
<p><strong>Paper Download Link (Source)</strong></p>
<p><strong><a href="https://www.researchpublish.com/papers/wavelet-attention-vgg19-and-xgboost-for-classification-of-skin-disease">https://www.researchpublish.com/papers/wavelet-attention-vgg19-and-xgboost-for-classification-of-skin-disease</a></strong></p>International Journal of Computer Science and Information Technology Research, ISSN 2348-1196 (print), ISSN 2348-120X (online), Research Publish Journals, Website: www.researchpublish.co
Chloroplast genome of Ecbolium viride: Plastome evolution and Phylogenomics of Justiceae (Acanthaceae, Acanthoideae)
No full textJusticieae is the most taxonomically complex tribe in Acanthaceae. Here, we sequenced the plastome of Ecbolium viride, a medicinally important species. The genome was analyzed with previously reported plastome of Justiceae. The plastome of E. viride has quadripartite structure with a length of 151, 185 bp. The comparative genomic analyses revealed no structural inversion in Justiceae and some regions (rpoC2, ycf2, ycf1 and ndhH rps16-trnQ-UGG, and trnL-CAA-ycf15) exhibiting a significant level of nucleotide divergence. The positive selection analyses revealed that some species in the tribe have undergone adaptive evolution. The visualization of the boundaries between the single copy and inverted repeat regions revealed that Justiceae chloroplast genome experienced some levels of variation which give an insight into the evolution of the species. The longest genome was in the earliest diverged taxa of the tribe P. haikangense and from this genome, a series of contraction and expansion occurred contributing to the evolution of other lineages. The plastome-based phylogeny revealed and confirmed the monophyly of Justiceae, polyphyly of Justicia and supported the tribal classification Graptophyllinae, Tetrameriinae, Isoglossinae. We proposed that Declipterinae should be treated as subtribe and the status of Justiciinae can only be confirmed after the resolution of the polyphyletic JusticiaThe presentation of the authors' names and (or) special characters in the title of the pdf file of the accepted manuscript may differ slightly from what is displayed on the item page. The information in the pdf file of the accepted manuscript reflects the original submission by the author
Electrical properties of graphene film for counter electrode in dye sensitized solar cells
No full textThe Concept of Moments of Order Invariant Quantum Processes.
No full textImportant information about stationary flows (BI)IÎI is contained in moments of the corresponding process. We consider the case where we have an order invariance of the moments of the increments, in the sense that such moments do not change if we shift the increments against each other, as long as we do not change the relative position of the intervals. We then show that for large classes of quantum processes one can make precise statements about the time behaviour of their moments. Keywords: Von Neumann algebras, Stationary flows and quantum processes, order invariant, Trace and moments
Problem of Integration with Respect to Unbounded Measures on the Set of Projections
No full textWe note if j is a normal weight on M, then is a measure on projections and if a measure on projections can be extended to a normal weight, then the problem of constructing an integral with respect to this measure reduces to the problem of constructing an integral with respect to the weight. We therefore present several methods of constructing noncommutative integration which gives a survey of the contemporary state of the theory in the von Neumann algebra (M) with respect to weightj. For every aÎ [0,1], the Banach space is isometrically isomorphic to the space Lp(t) and the space is, by definition, the Banach space completion of in the norm .We construct the scale of Lp(j) spaces with respect to a faithful normal semifinite (f.n.s.) weight j on a von Neumann algebra M. These spaces are realized by operators. This is achieved by extending the original algebra M, and the Hilbert space where M originally acted is altered, as well. In the construction of the scale, the concept of an operator-valued weight is used. We discuss the problem of integration with respect to measures on projections which remains open for unbounded measures (m(1) = +¥) and their structure has been studied only for the algebra ?(?). Keywords: Von Neumann algebra, Faithful normal semifinite trace (f.n.s.)t, weight, isometrically isomorphic, projections, Banach spaces and Lp-space
A Framework for Facilitating Secure Design and Development of IoT Systems
No full textThe term Internet of Things (IoT) describes an ever-growing ecosystem of physical objects
or things interconnected with each other and connected to the Internet. IoT devices
consist of a wide range of highly heterogeneous inanimate and animate objects. Thus, a
thing in the context of the IoT can even mean a person with blood pressure or heart rate
monitor implant or a pet with a biochip transponder. IoT devices range from ordinary
household appliances, such as smart light bulbs or smart coffee makers, to sophisticated
tools for industrial automation. IoT is currently leading a revolutionary change in many
industries and, as a result, a lot of industries and organizations are adopting the paradigm
to gain a competitive edge. This allows them to boost operational efficiency and optimize
system performance through real-time data management, which results in an optimized
balance between energy usage and throughput. Another important application area is
the Industrial Internet of Things (IIoT), which is the application of the IoT in industrial
settings. This is also referred to as the Industrial Internet or Industry 4.0, where Cyber-
Physical Systems (CPS) are interconnected using various technologies to achieve wireless
control as well as advanced manufacturing and factory automation. IoT applications
are becoming increasingly prevalent across many application domains, including smart
healthcare, smart cities, smart grids, smart farming, and smart supply chain management.
Similarly, IoT is currently transforming the way people live and work, and hence
the demand for smart consumer products among people is also increasing steadily. Thus,
many big industry giants, as well as startup companies, are competing to dominate the
market with their new IoT products and services, and hence unlocking the business value
of IoT.
Despite its increasing popularity, potential benefits, and proven capabilities, IoT is still in
its infancy and fraught with challenges. The technology is faced with many challenges, including
connectivity issues, compatibility/interoperability between devices and systems,
lack of standardization, management of the huge amounts of data, and lack of tools for
forensic investigations. However, the state of insecurity and privacy concerns in the IoT
are arguably among the key factors restraining the universal adoption of the technology.
Consequently, many recent research studies reveal that there are security and privacy issues
associated with the design and implementation of several IoT devices and Smart Applications
(smart apps). This can be attributed, partly, to the fact that as some IoT device
makers and smart apps development companies (especially the start-ups) reap business
value from the huge IoT market, they tend to neglect the importance of security. As a
result, many IoT devices and smart apps are created with security vulnerabilities, which
have resulted in many IoT related security breaches in recent years.
This thesis is focused on addressing the security and privacy challenges that were briefly
highlighted in the previous paragraph. Given that the Internet is not a secure environ ment even for the traditional computer systems makes IoT systems even less secure due
to the inherent constraints associated with many IoT devices. These constraints, which are
mainly imposed by cost since many IoT edge devices are expected to be inexpensive and
disposable, include limited energy resources, limited computational and storage capabilities,
as well as lossy networks due to the much lower hardware performance compared
to conventional computers. While there are many security and privacy issues in the IoT
today, arguably a root cause of such issues is that many start-up IoT device manufacturers
and smart apps development companies do not adhere to the concept of security by
design. Consequently, some of these companies produce IoT devices and smart apps with
security vulnerabilities.
In recent years, attackers have exploited different security vulnerabilities in IoT infrastructures
which have caused several data breaches and other security and privacy incidents
involving IoT devices and smart apps. These have attracted significant attention
from the research community in both academia and industry, resulting in a surge of proposals
put forward by many researchers. Although research approaches and findings may
vary across different research studies, the consensus is that a fundamental prerequisite for
addressing IoT security and privacy challenges is to build security and privacy protection
into IoT devices and smart apps from the very beginning. To this end, this thesis investigates
how to bake security and privacy into IoT systems from the onset, and as its main
objective, this thesis particularly focuses on providing a solution that can foster the design
and development of secure IoT devices and smart apps, namely the IoT Hardware Platform
Security Advisor (IoT-HarPSecA) framework. The security framework is expected to
provide support to designers and developers in IoT start-up companies during the design
and implementation of IoT systems. IoT-HarPSecA framework is also expected to facilitate
the implementation of security in existing IoT systems.
To accomplish the previously mentioned objective as well as to affirm the aforementioned
assertion, the following step-by-step problem-solving approach is followed. The first step
is an exhaustive survey of different aspects of IoT security and privacy, including security requirements in IoT architecture, security threats in IoT architecture, IoT application domains
and their associated cyber assets, the complexity of IoT vulnerabilities, and some
possible IoT security and privacy countermeasures; and the survey wraps up with a brief
overview of IoT hardware development platforms. The next steps are the identification of
many challenges and issues associated with the IoT, which narrowed down to the abovementioned
fundamental security/privacy issue; followed by a study of different aspects of
security implementation in the IoT. The remaining steps are the framework design thinking
process, framework design and implementation, and finally, framework performance
evaluation.
IoT-HarPSecA offers three functionality features, namely security requirement elicitation security best practice guidelines for secure development, and above all, a feature that recommends
specific Lightweight Cryptographic Algorithms (LWCAs) for both software and
hardware implementations. Accordingly, IoT-HarPSecA is composed of three main components,
namely Security Requirements Elicitation (SRE) component, Security Best Practice
Guidelines (SBPG) component, and Lightweight Cryptographic Algorithms Recommendation
(LWCAR) component, each of them servicing one of the aforementioned features.
The author has implemented a command-line tool in C++ to serve as an interface
between users and the security framework. This thesis presents a detailed description,
design, and implementation of the SRE, SBPG, and LWCAR components of the security
framework. It also presents real-world practical scenarios that show how IoT-HarPSecA
can be used to elicit security requirements, generate security best practices, and recommend
appropriate LWCAs based on user inputs. Furthermore, the thesis presents performance
evaluation of the SRE, SBPG, and LWCAR components framework tools, which
shows that IoT-HarPSecA can serve as a roadmap for secure IoT development.O termo Internet das coisas (IoT) é utilizado para descrever um ecossistema, em expansão,
de objetos físicos ou elementos interconetados entre si e à Internet. Os dispositivos
IoT consistem numa gama vasta e heterogénea de objetos animados ou inanimados e,
neste contexto, podem pertencer à IoT um indivíduo com um implante que monitoriza a
frequência cardíaca ou até mesmo um animal de estimação que tenha um biochip. Estes
dispositivos variam entre eletrodomésticos, tais como máquinas de café ou lâmpadas inteligentes,
a ferramentas sofisticadas de uso na automatização industrial. A IoT está a
revolucionar e a provocar mudanças em várias indústrias e muitas adotam esta tecnologia
para incrementar as suas vantagens competitivas. Este paradigma melhora a eficiência
operacional e otimiza o desempenho de sistemas através da gestão de dados em tempo
real, resultando num balanço otimizado entre o uso energético e a taxa de transferência.
Outra área de aplicação é a IoT Industrial (IIoT) ou internet industrial ou Indústria 4.0,
ou seja, uma aplicação de IoT no âmbito industrial, onde os sistemas ciberfísicos estão interconectados
a diversas tecnologias de forma a obter um controlo de rede sem fios, bem
como fabricações avançadas e automatização fabril. As aplicações da IoT estão a crescer
e a tornarem-se predominantes em muitos domínios de aplicação inteligentes como sistemas
de saúde, cidades, redes, agricultura e sistemas de fornecimento. Da mesma forma,
a IoT está a transformar estilos de vida e de trabalho e assim, a procura por produtos inteligentes
está constantemente a aumentar. As grandes indústrias e startups competem
entre si de forma a dominar o mercado com os seus novos serviços e produtos IoT, desbloqueando
o valor de negócio da IoT.
Apesar da sua crescente popularidade, benefícios e capacidades comprovadas, a IoT está
ainda a dar os seus primeiros passos e é confrontada com muitos desafios. Entre eles,
problemas de conectividade, compatibilidade/interoperabilidade entre dispositivos e sistemas,
falta de padronização, gestão das enormes quantidades de dados e ainda falta de
ferramentas para investigações forenses. No entanto, preocupações quanto ao estado de
segurança e privacidade ainda estão entre os fatores adversos à adesão universal desta
tecnologia. Estudos recentes revelaram que existem questões de segurança e privacidade
associadas ao design e implementação de vários dispositivos IoT e aplicações inteligentes
(smart apps.), isto pode ser devido ao facto, em parte, de que alguns fabricantes e empresas
de desenvolvimento de dispositivos (especialmente startups) IoT e smart apps., recolham
o valor de negócio dos grandes mercados IoT, negligenciando assim a importância
da segurança, resultando em dispositivos IoT e smart apps. com carências e violações de
segurança da IoT nos últimos anos.
Esta tese aborda os desafios de segurança e privacidade que foram supra mencionados.
Visto que a Internet e os sistemas informáticos tradicionais são por vezes considerados inseguros,
os sistemas IoT tornam-se ainda mais inseguros, devido a restrições inerentes a tais dispositivos. Estas restrições são impostas devido ao custo, uma vez que se espera que
muitos dispositivos de ponta sejam de baixo custo e descartáveis, com recursos energéticos
limitados, bem como limitações na capacidade de armazenamento e computacionais,
e redes com perdas devido a um desempenho de hardware de qualidade inferior, quando
comparados com computadores convencionais. Uma das raízes do problema é o facto
de que muitos fabricantes, startups e empresas de desenvolvimento destes dispositivos e
smart apps não adiram ao conceito de segurança por construção, ou seja, logo na conceção,
não preveem a proteção da privacidade e segurança. Assim, alguns dos produtos e
dispositivos produzidos apresentam vulnerabilidades na segurança.
Nos últimos anos, hackers maliciosos têm explorado diferentes vulnerabilidades de segurança
nas infraestruturas da IoT, causando violações de dados e outros incidentes de
privacidade envolvendo dispositivos IoT e smart apps. Estes têm atraído uma atenção significativa
por parte das comunidades académica e industrial, que culminaram num grande
número de propostas apresentadas por investigadores científicos. Ainda que as abordagens
de pesquisa e os resultados variem entre os diferentes estudos, há um consenso e
pré-requisito fundamental para enfrentar os desafios de privacidade e segurança da IoT,
que buscam construir proteção de segurança e privacidade em dispositivos IoT e smart
apps. desde o fabrico. Para esta finalidade, esta tese investiga como produzir segurança
e privacidade destes sistemas desde a produção, e como principal objetivo, concentra-se
em fornecer soluções que possam promover a conceção e o desenvolvimento de dispositivos
IoT e smart apps., nomeadamente um conjunto de ferramentas chamado Consultor
de Segurança da Plataforma de Hardware da IoT (IoT-HarPSecA). Espera-se que o conjunto
de ferramentas forneça apoio a designers e programadores em startups durante a
conceção e implementação destes sistemas ou que facilite a integração de mecanismos de
segurança nos sistemas préexistentes.
De modo a alcançar o objetivo proposto, recorre-se à seguinte abordagem. A primeira fase
consiste num levantamento exaustivo de diferentes aspetos da segurança e privacidade na
IoT, incluindo requisitos de segurança na arquitetura da IoT e ameaças à sua segurança,
os seus domínios de aplicação e os ativos cibernéticos associados, a complexidade das
vulnerabilidades da IoT e ainda possíveis contramedidas relacionadas com a segurança e
privacidade. Evolui-se para uma breve visão geral das plataformas de desenvolvimento
de hardware da IoT. As fases seguintes consistem na identificação dos desafios e questões
associadas à IoT, que foram restringidos às questões de segurança e privacidade. As demais
etapas abordam o processo de pensamento de conceção (design thinking), design e
implementação e, finalmente, a avaliação do desempenho.
O IoT-HarPSecA é composto por três componentes principais: a Obtenção de Requisitos
de Segurança (SRE), Orientações de Melhores Práticas de Segurança (SBPG) e a recomendação
de Componentes de Algoritmos Criptográficos Leves (LWCAR) na implementação de software e hardware. O autor implementou uma ferramenta em linha de comandos
usando linguagem C++ que serve como interface entre os utilizadores e a IoT-HarPSecA.
Esta tese apresenta ainda uma descrição detalhada, desenho e implementação das componentes
SRE, SBPG, e LWCAR. Apresenta ainda cenários práticos do mundo real que
demostram como o IoT-HarPSecA pode ser utilizado para elicitar requisitos de segurança,
gerar boas práticas de segurança (em termos de recomendações de implementação) e recomendar
algoritmos criptográficos leves apropriados com base no contributo dos utilizadores.
De igual forma, apresenta-se a avaliação do desempenho destes três componentes,
demonstrando que o IoT-HarPSecA pode servir como um roteiro para o desenvolvimento
seguro da IoT.Project through FCT/COMPETE/FEDER under Grant POCI01-0145-FEDER-030657, in part by the FCT/MCTES through national funds and, when
applicable, co-funded EU funds under the project UIDB/50008/2020, and in part by the
Fundação para a Ciência e Tecnologia (FCT) research grant with reference BIM/n◦32/2018-
B00582
An estimation of temperature in living tissue using a fractional model with sinusoidal heat flux conditions on the skin surface
No full textAn accurate and quantitative description of the thermal responses of skin tissue under sinusoidal heat flux conditions on the skin surface is provided in this study using a fractional model of the bioheat transfer equation. The governing fractional one-dimensional bio-heat transfer equation was transformed into a dimensionless form to get closed solution by the Caputo-Fabrizio time-fractional derivative technique. The proposed technique, with the aid of symbolic computation, provides an impressive solution for the bioheat transfer equation. Numerical simulations were performed with the aid of Mathcad software to study the behavior of temperature transients on the skin surface exposed to instantaneous surface heating. To investigate the thermal impacts of various control factors on tissue temperature, sensitivity analysis is carried out. We found that the effects of the fractional derivative and the moving heat source velocity on the temperature of the tissue and thermal injuries are enormous, and are shown graphically for an explicit and detail discussion. Observations from the graphical representation of results, show clearly that increasing pressure term leads to increase in velocity distribution of temperature and the effect is maximum towards the center in most cases. The study reveals that bio-thermal research can help with skin burn evaluation, clinical thermal treatment equipment design, and thermal protection from various risks of skin heat injuries. In conclusion, the results of this investigation will be very helpful and useful to both clinical-therapeutic applications and medical sciences
PEMBANGUNAN SERTA PERBAIKAN FASILITAS UMUM (KAMAR MANDI, WATER CLOSET DAN SEPTIC TANK) DI KELURAHAN SUPRAU DISTRIK MALADUM MES
No full textDistrik Maladum Mes, merupakan sebuah Distrik dengan lokasi yang cukup strategis. Distrik Maladum Mes berada di wilayah Pemerintahan Kota Sorong dan memiliki luas126,40 Km2. Kelurahan Suprau memiliki lahan dataran yang rendah dan pesisir pantai yang cukup luas untuk sektor perikanan dan nelayan. Hal ini dibuktikan dengan letak kelurahan yang berada dipinggiran pantai. Sebagian besar mata pencarian penduduk Kelurahan Suprau di dominasi oleh nelayan dan pedagang kaki lima. Profesi penduduk di lingkungan kelurahan Suprau sangat beragam. Sebagian besar penduduk di lingkungan kelurahan Suprau berprofesi sebagai nelayan, pedagang kaki lima, PNS dan karyawan swasta. Program Pengabdian Masyarakat yang dilakukan kali ini memiliki sasaran langsung kepada warga Suprau. Mitra dalam pengabdian ini adalah Kelurahan Suprau. Mitra dipertimbangkan dengan melihat kondisi dan situasi selama survey, dimana wilayah ini kurang diperhatikan oleh pemerintah daerah, dan kerusakan – kerusakan yang ada merupakan fasilitas umum yang sangat diperlukan oleh warga suprau. Sebagian besar masyarkat suprau menggunakan kamar mandi dan tolilet/WC secara bersama – sama. Untuk itu dalam pengabdian kali ini kami menitikberatkan dalam pembangunan fasilitas yang dibutuhkan oleh masyarakat. kelurahan Suprau
A Modification of Conjugate Gradient Parameter and Its Global Convergence for Solving Unconstrained Optimization Problems
No full textThe methods of nonlinear conjugate gradient coefficients are significant and helpful in solving large-scale unconstrained optimization problems, due to their simplicity and lower storage requirement. Research activities on its applications to handle higher-dimensional systems of nonlinear equations are of paramount importance. Many authors studied and developed different kinds of conjugate gradient coefficients. Recently, some conjugate gradients were proposed with large dimensions, but they have small sample sizes. So, they cannot solve problems that have higher dimensions. Therefore, this research proposed a modified algorithm with a huge dimension and a large sample size. The strategy of strong Wolfe line search was applied in the convergence analysis, which makes it possible to converge globally with a descent property. Finally, the numerical results show that the proposed algorithm performs more efficiently and is superior to the existing Conjugate Gradient (CG) coefficients
Systematic review and meta-analysis of tuberculosis in animals in Nigeria
No full textAnimal tuberculosis (TB) is a contagious and chronic disease caused by mycobacteria belonging to theMycobacterium tuberculosis complex (MTBC) in domestic and wild animals. MTBC strains infection has been confirmed in many animal species in Nigeria, including captive wildlife, cattle, dromedary camels, goats, and pigs. Despite widespread infection and the potential impact of the disease on public health, active surveillance and control strategies are absent in Nigeria. This study aimed to conduct the first comprehensive meta-analysis to assess the distribution of tuberculosis and analyze the potential moderators of infection in animals in Nigeria. Eligible studies (sixty-one (Cadmus et al., 2014) [61] prevalence and seven (Menzies and Neill, 2000) [7] case reports) were retrieved and included in the analysis. The analyses showed an overall pooled TB prevalence of 7.0% (95% CI: 6.0-8.0) comprising of infection distributed in cattle (8.0%, 95% CI: 7.0-8.0), goats (0.47%, 95% CI: 0-1.2), sheep (0.27%, 95% CI: 0.14-0.46), camels (13.0%, 95% CI: 0-47), and wildlife (13.0%, 95% CI: 9-16) respectively. The occurrence of infection was significantly moderated by the publication periods, geographical location, sample size, and detection methods. TB prevalence was heterogeneous across several predictors, with the year of publication exhibiting a higher rate (46%) of the detected heterogeneity. These findings should provide policy-relevant information to guide the design and establishment of prevention and control measures amenable to the local situations in Nigeria. © 2023 The Author(s
