Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

31,456 Article Results

Mortality prediction of COVID-19 patients using supervised machine learning

10.11591/ijai.v13.i4.pp4472-4479
Husnul Khuluq , Prasandhya Astagiri Yusuf , Dyah Aryani Perwitasari , Thang Nguyen
Hospitalized patients with COVID-19 are at higher risk of mortality. Machine learning (ML) algorithms have been proposed as a possible strategy for predicting mortality rates among patients hospitalized with COVID-19. This study analyzed various ML algorithms and identified the best model to predict COVID-19 mortality based on demographic, clinical, and laboratory data collected at registration. Data from 4,314 eligible patients (3,384 survivors and 930 who died) was collected from the register of three hospitals in Yogyakarta province, Indonesia, based on the confirmed predictors. Next, ML algorithms were utilized to predict mortality. Finally, the confusion matrix was used to evaluate how effective the models performed. The best five predictors from 26 features were myocardial infarction, SpO2, neutrophil, D dimer, and creatinine. The results indicate that the random forest algorithm showed better performance than other ML algorithms in terms of accuracy, sensitivity, precision, specificity, and area under the curve (AUC), achieving values of 84.15%, 84.0%, 84.1%, 83.9%, and 90.02%, respectively. Implementing ML techniques can accurately predict the mortality rate associated with COVID-19. Therefore, this predictive model can help clinicians and hospitals predict COVID patients with a greater risk of death and effectively target more appropriate treatments.
Volume: 13
Issue: 4
Page: 4472-4479
Publish at: 2024-12-01

DeepCervix: enhancing cervical cancer detection through transfer learning with VGG-16 architecture

10.11591/ijeecs.v36.i3.pp1895-1903
Vaishali M. Joshi , Prajkta P. Dandavate , R. Rashmi , Gitanjali R. Shinde , Neeta N. Thune , Riddhi Mirajkar
Cervical cancer remains a significant global health concern, emphasizing the urgent need for improved detection methods to ensure timely treatment. This research introduces a sophisticated methodology leveraging recent advances in medical imaging and deep learning algorithms to enhance the accuracy and efficiency of cervical cancer detection. The proposed approach comprises meticulous data preprocessing to ensure the integrity of input images, followed by the training of deep learning models including ResNet-50, AlexNet, and VGG-16, renowned for their performance in computer vision tasks. Evaluation metrics such as accuracy, precision, recall, and F1-score demonstrate the efficacy of the methodology, with an outstanding accuracy rate of 98% achieved. The model’s proficiency in accurately distinguishing healthy cervical tissue from cancerous tissue is underscored by precision, recall, and F1-score values. The primary strength of this deep learning-based approach lies in its potential for early detection, promising significant impact on cervical cancer diagnosis and treatment outcomes. This methodology contributes to advancements in medical imaging techniques, facilitating improved outcomes in cervical cancer detection and treatment.
Volume: 36
Issue: 3
Page: 1895-1903
Publish at: 2024-12-01

Enhanced scene text recognition using deep learning based hybrid attention recognition network

10.11591/ijai.v13.i4.pp4927-4938
Ratnamala S Patil , Geeta Hanji , Rakesh Huded
The technique of automatically recognizing and transforming text that is present in pictures or scenes into machine-readable text is known as scene text recognition. It facilitates applications like content extraction, translation, and text analysis in real-world visual data by enabling computers to comprehend and extract textual information from images, videos, or documents. Scene text recognition is essential for many applications, such as language translation and content extraction from photographs. The hybrid attention recognition network (HARN), unique technology presented in this research, is intended to greatly improve efficiency and accuracy of text recognition in complicated scene situations. HARN makes use of cutting-edge elements including alignment-free sequence-to-sequence (AFS) module, creative attention mechanisms, and hybrid architecture that blends attention models with convolutional neural networks (CNNs). Thanks to its novel attention processes, HARN is capable of comprehending wide range of scene text components by capturing both local and global context information. Through faster network convergence, shorter training times, and better utilization of computing resources, the suggested technique raises bar for state-of-the-art. HARN’s versatility makes it a good choice for range of scene text recognition applications, including multilingual text analysis and data extraction. Extensive tests are conducted to assess the effectiveness of HARN approach and demonstrate it is ability to greatly influence real-world applications where accurate and efficient text recognition is essential.
Volume: 13
Issue: 4
Page: 4927-4938
Publish at: 2024-12-01

Energy baseline model enhanced based on artificial neural network in industrial buildings

10.11591/ijeecs.v36.i3.pp1493-1502
Younes Ouaomar , Said Benkachcha , Mourad Kaddiri
In this article, a new energy-efficient reference model has been established for a plastic injection molding plant. However, the proposed model handled difficulties due to the lack of robust and complete data, such as production mix and cooling degree-days. In addition, the proposed model applies three distinct enhanced modeling methodologies, including regression modeling, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS). Furthermore, these performance parameters were established to assess the accuracy of each model in this work. Moreover, the numerical results show that among the methodologies used in this work, the ANN demonstrated effective performance despite uncertainties in the measured input variables. The ANN numerical results in this paper highlight the ability to accurately assess baseline consumption in the industrial sector, providing a practical tool for decision-makers to improve energy efficiency.
Volume: 36
Issue: 3
Page: 1493-1502
Publish at: 2024-12-01

Impact study on indirect lightning strikes on photovoltaic systems near transmission lines

10.12928/telkomnika.v22i6.26613
Anis Niza; Universiti Teknikal Malaysia Melaka Ramani , Rayyan Che; Suria & Sonne Sdn. Bhd Razali , Aine Izzati; Universiti Teknikal Malaysia Melaka Tarmizi
Grid-integrated photovoltaic (PV) systems are currently undergoing explosive growth in Malaysia. However, as more PV systems are installed close to transmission lines, there are concerns about the impact of electromagnetic (EM) properties affecting the performance and operation of the PV systems if they are exposed to lightning strikes, particularly indirect ones. Therefore, this study aims to model the impact of indirect lightning strikes on PV systems installed in proximity to transmission lines. The model involves developing a 3D model of the PV system together with a sample transmission line and creating an artificial lightning event to study the EM activity within the area. The results are compared to the IEC 61000 standard to determine its level of r\hazard. From the simulation results, it was found that the current intensity, H, of the disturbance, can reach up to 63% more than the standard limit stipulated in the standards. The significance of the study ensures that PV systems installed within the vicinity of power lines or substations have adequate lightning protection systems as well as proper earthing systems.
Volume: 22
Issue: 6
Page: 1581-1589
Publish at: 2024-12-01

Entanglement classification – a comparative study of 𝑺𝑼(𝟐) and 𝑺𝑳(𝟐) developed operator model

10.11591/ijict.v13i3.pp556-562
Amirul Asyraf Zhahir , Siti Munirah Mohd , Mohd Ilias M. Shuhud , Bahari Idrus , Hishamuddin Zainuddin , Nurhidaya Mohamad Jan , Mohamed Ridza Wahiddin
Entanglement classification is a core aspect of quantum information theory. It ensures successful quantum information processing. This article presents a comparative study of entanglement classification using developed operator models for the special unitary group and special linear group. This study was built upon prior work in entanglement classification in a pure three-qubit quantum system environment, where the operator models for each mathematical group were independently developed. Through extensive analysis, both synthesized models are functionally effective and yield the desired results. However, the comparative analysis reveals that the operator model exhibits certain limitations, particularly in its early phase of development compared to. This study provides significant enlightenments into the practical abilities of the developed operator models in entanglement classification and underlines the theoretical distinction between and paving the path for future research in quantum information theory, specifically entanglement classification.
Volume: 13
Issue: 3
Page: 556-562
Publish at: 2024-12-01

Enhancing flexural resistance in hot mix asphalt: a study of effects of wire mesh on load-bearing capacity

10.11591/ijaas.v13.i4.pp1043-1049
Romaynoor Ismy , Suhaimi Suhaimi , Raden Dedi Iman Kurnia , Richard Mareno
In transportation engineering, road pavement is commonly categorized as either flexible pavement or rigid pavement. The pavement demonstrates a distinct capacity to endure a variety of loads, including compressive and tensile loads. The capacity to endure compressive and tensile forces is extremely important, especially in the field of pavement construction, as it ensures both the longevity and the safety of the pavement. The objective of this study was to evaluate the capacity of hot mix asphalt to endure compressive and tensile pressures. The experimental methodology employed four different wire mesh deployment configurations on hot asphalt mixtures, utilizing three-point flexural test equipment. The data indicates the most effective method for mimicking hot mix asphalt involves adding a wire mesh layer at a depth of 30 mm below the surface of the experimental specimen. The particular modeling method showed a measurement of flexural resistance up to 291.85 kN. The study's findings indicate that the hot asphalt mixture exhibits a state of balance in its capacity to endure both compressive and tensile pressures. Incorporating a wire mesh layer within the middle section of the hot asphalt mixture has been perceived to enhance its ability to withstand tensile loads, hence improving its overall performance.
Volume: 13
Issue: 4
Page: 1043-1049
Publish at: 2024-12-01

Automated blood cancer detection models based on EfficientNet-B3 architecture and transfer learning

10.11591/ijeecs.v36.i3.pp1731-1738
Nawaf Alshdaifat , Hamza Abu Owida , Zaid Mustafa , Ahmad Aburomman , Suhaila Abuowaida , Abdullah Ibrahim , Wafa Alsharafat
In blood smear images, there are difficulties in diagnosing blood cancer diseases like leukemia and lymphoma because of their various forms that appear in the human body. In this paper, a method for automatic detection of blood cancer is suggested that uses the EfficientNet-B3 architecture along with transfer learning techniques to improve accuracy and efficiency. We first fine-tuned the EfficientNet-B3 model, which was pre-trained on a large dataset consisting of annotated blood smear images, to capture pertinent features linked with blood malignant cells. To expedite the training process and adapt the model to our task, we use transfer learning. The proposed approach’s results from our experiments show that it outperforms traditional deep learning models and state-of-the-art methods in blood cancer detection. Additionally, with high precision and recall rates, this model also detects different types of blood cancers with robustness in its performance since its accuracy is over 99%. This means that when used together with the EfficientNet-B3 architecture, transfer learning can help the developed methods generalize among different types of blood cancers and conditions.
Volume: 36
Issue: 3
Page: 1731-1738
Publish at: 2024-12-01

Digital learning using ChatGPT in elementary school mathematics learning: a systematic literature review

10.11591/ijeecs.v36.i3.pp1701-1710
Prabandari Listyaningrum , Heri Retnawati , Harun Harun , Hamidulloh Ibda
Digital learning with ChatGPT in elementary school mathematics learning is urgently implemented. Several studies have explored digital learning with ChatGPT in elementary school mathematics learning, but studies using SLR are minimal. This article presents the 2022-2024 SLR study on digital learning with ChatGPT in elementary school mathematics learning. This SLR and PRISMA method is supported by Publish or Perish 8, VOSviewer version 1.6.20, Mendeley version 1.19.8, and ATLAS.ti version 7.5.16. The search results obtained 1,259 Scopus articles, which were filtered to 40 and analyzed using ATLAS.ti, then the results were described according to the research question. Digital learning with ChatGPT is a learning approach using the synchronous-asynchronous mode, virtual classrooms, distance, use of interactive digital tools, digital methods and media, innovation, digital modeling, use of robotics and AI ChatGPT for children with the principle of collaboration digital, and problem-solving with the support of digital resources. ChatGPT features multilingual, natural language, advanced AI, 24/7 availability, answering math questions, recurring training, and helping students with various math tasks. Implementation of digital learning with ChatGPT in elementary school mathematics learning for problem-solving, geometry, function limits in algebra, the material on flat shapes, geometric shapes, integrated PjBL, online, mixed and flipped classes.
Volume: 36
Issue: 3
Page: 1701-1710
Publish at: 2024-12-01

Enhancing energy efficiency with poncebloc and hemp wool: a Moroccan case study

10.11591/ijaas.v13.i4.pp974-984
Yasser Boukioud , Kaoutar Senhaji Rhazi , Youssef Mejdoub
In recent years, the building sector's energy consumption in Morocco has been steadily increasing, highlighting the urgent need for sustainable building practices. Optimizing building parameters and the development of new insulation materials have become critical areas of research due to the significant energy losses associated with traditional insulation materials. Insulation materials can be a simple and cost-effective way to improve building energy efficiency and reduce environmental impact. Unfortunately, not all of these materials exhibit favorable thermal characteristics such as conductivity or heat capacity. Some involve complex production processes, while others are prohibitively expensive or toxic. At this juncture, the strategy of integrating ecological materials with insulation materials to enhance thermal performance holds considerable promise. This paper investigates the impact of the integration of poncebloc and hemp wool in inside temperature and annual consumption of a building construction numerically under a transient simulation program (TRNSYS). The results show that by using these two materials along with double-glazed windows, energy savings ranging from 50-55% can be achieved.
Volume: 13
Issue: 4
Page: 974-984
Publish at: 2024-12-01

Hybrid load forecasting considering energy efficiency and renewable energy using neural network

10.11591/ijaas.v13.i4.pp763-772
Adriana Haziqah Mohd Aizam , Nofri Yenita Dahlan , Saidatul Habsah Asman , Siti Hajar Yusoff
In recent years, the relationship between a country's gross domestic product (GDP) and its electricity consumption has changed significantly due to increased energy efficiency (EE) and renewable energy (RE) adoption. This decoupling disrupts conventional load forecasting models, affecting utility companies. This study has developed an innovative solution using an artificial neural network (ANN) Hybrid method for load forecasting, resulting in a remarkably accurate model with 99.68% precision. Applying this model to Malaysia's electricity consumption from 2020 to 2040 reveals a significant 13% reduction when accounting for EE and RE trends. This method aids risk management, contingency planning, and decision-making by accurately reflecting changing energy usage dynamics influenced by EE and RE sources.
Volume: 13
Issue: 4
Page: 763-772
Publish at: 2024-12-01

A new hybrid LED driver

10.11591/ijpeds.v15.i4.pp2367-2375
Orazbayev Sagi , Dzhunusbekov Erlan
The ongoing challenge in light-emitting diode (LED) lighting is to integrate power and control circuits into a single chip to reduce the number and cost of components, minimize size, and enhance reliability. This integration requires a reduction in power dissipation on the chip, improved efficiency, and smaller magnetic components. Recently, hybrid LED drivers have been proposed to address this integration challenge. Unlike traditional transformerless single-stage LED drivers, which implement boost or buck topologies, hybrid drivers reduce pulsed voltage on power semiconductors. The proposed hybrid LED driver aims to prevent power switches from hard switching, which in turn reduces metal–oxide semiconductor field-effect transistor or MOSFETs' overheating in the integrated circuit body, increases the reliability of the chip, and alleviates electromagnetic interference or EMI issues. The proposed improvement to the hybrid driver is analyzed. The control method was suggested. Computer modelling was provided to confirm the feasibility, controllability and resulting advantages of the proposal. The proposed hybrid LED driver is attractive for circuit integration to reduce size and cost compared to the achievements of previous art.
Volume: 15
Issue: 4
Page: 2367-2375
Publish at: 2024-12-01

Correlation of physical activity with cognition and mental health in medical students

10.11591/ijphs.v13i4.23150
Yhusi Karina Riskawati , Faresa Aulia Rastramadhani , Henny Mulyani , Ridha Saniyyah Rindradi , Muhammad Isa Mahendra , Christyaji Indradmojo , Aswaty Nur
Low physical activity increases the number of non-communicable diseases. This phenomenon occurs in medical students. This study aimed to determine the correlation between sedentary behavior and physical activity on cognition, mental health, cortisol, and brain derived neurotropic factor (BDNF). Using a cross-sectional design of observational study, eighty-six medical students were involved in this study. The subjects were interviewed to assess cognitive function and depression levels. Saliva was taken to measure cortisol and BDNF level. Bivariate analysis was performed using the Spearman test. Depression is the only variable that correlates significantly with habitual physical activity (p=0.025, r=-0.214). Sedentary behavior has a weak correlation with cognitive failure, anxiety, and depression ([p=0.046, r=0.216]; [p=0.039, r=0.223]; [p=0.011, r=0.273]). The results found that high physical activity improves mental health and cognition. This study suggested that physical activity can alleviate symptoms of depression and anxiety among medical students.
Volume: 13
Issue: 4
Page: 1617-1625
Publish at: 2024-12-01

Survey of IoT and AI applications: future challenges and opportunities in agriculture

10.11591/ijeecs.v36.i3.pp1655-1663
Kamal Elhattab , Said Elatar
The internet of things (IoT) connects physical objects through sensors, software, and communication technologies, enabling efficient data collection and sharing. This interconnection promotes automation, real-time monitoring, and improved decision-making across various sectors. In agriculture, the integration of IoT with artificial intelligence (AI) is revolutionizing resource management by providing farmers with real-time information on crop health, climate conditions, and soil quality. This paper explores how IoT and AI are transforming traditional agricultural practices to enhance both efficiency and sustainability. Through an in-depth analysis of existing literature and practical applications in the sector, this study identifies significant advancements in crop management, reduction of losses, and resource optimization. Additionally, it highlights persistent challenges such as data security and interoperability. The aim is to address these challenges and propose innovative solutions to optimize agricultural processes. The results indicate that while IoT and AI offer substantial benefits, further advancements and solutions are needed to fully leverage these technologies for sustainable agricultural development.
Volume: 36
Issue: 3
Page: 1655-1663
Publish at: 2024-12-01

Deep learning approach for forensic facial reconstruction depends on unidentified skull

10.11591/ijai.v13.i4.pp3858-3868
Doaa M. Mohammed , Mostafa Elgendy , Mohamed Taha
Facial reconstruction, or facial approximation, is an essential problem in a criminal investigation involving reconstructing a victim's face from his skull to determine the victim's identification at a crime scene. Facial approximation plays a crucial part when there is a lack of clues with investigators. Investigators utilize facial approximation to guess the victims' identities. This research attempted to use computer-aided face reconstruction rather than traditional approaches. Traditional methods of face reconstruction include the use of clay or gypsum. Traditional procedures necessitate forensic professionals to rebuild the victim's face. This research uses the convolution neural network skull part with sift (CNNSPS) model is employed to reconstruct facial features from a skull image utilizing public datasets CelebAMask-HQ and MUG500+. The proposed algorithm was tested on unidentified skull databases, and celebrity faces were used. The genuine datasets are not available, which is the key issue in this research.
Volume: 13
Issue: 4
Page: 3858-3868
Publish at: 2024-12-01
1 2 3
...
2097 2098
Show 388 of 2098

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration