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

29,939 Article Results

An optimized transfer learning-based approach for Crocidolomia pavonana larvae classification

10.11591/ijai.v14.i3.pp2270-2281
Risnawati Risnawati , Rodiah Rodiah , Sarifuddin Madenda , Diana Tri Susetianingtias
The increasing demand for mustard greens has driven farmers to continuously improve mustard greens cultivation. One of the challenges in mustard greens cultivation is the presence of insect pests. A significant pest in mustard greens is Crocidolomia pavonana (C. pavonana). C. pavonana damages plants by feeding on various parts, especially the leaves. The initial step in controlling them is insect pest monitoring. Monitoring aims to establish the control threshold. C. pavonana larvae have four instar stages: instar 1, 2, 3, and 4. Identification of the instar larval stages utilizes deep convolutional neural network (CNN) to classify C. Pavonana larvae on mustard greens using ResNet50V2 and DenseNet169 architectures optimized to enhance classification accuracy. The classification evaluation results show that both DenseNet169 and ResNet50V2 models achieve high accuracy, with DenseNet169 reaching the highest accuracy at 97.1%, while ResNet50V2 achieves an accuracy of 94.2%. The lower loss values on the test data compared to the validation data indicate that the deep learning models have successfully captured the patterns in C. pavonana images for classification. This classification process is expected to be one of the activities in monitoring the instar larvae to improve the accuracy of insecticide spraying and enhance mustard greens production.
Volume: 14
Issue: 3
Page: 2270-2281
Publish at: 2025-06-01

Developing gallium nitride-based inverters for high-performance photovoltaic integration in alternating current grids

10.11591/ijece.v15i3.pp2583-2598
Arsalan Muhammad Soomar , Piotr Musznicki
This study introduces a gallium nitride (GaN) based inverter optimized for alternating current (AC) grid integration, featuring a novel phase-locked loop (PLL) controller enhanced with sliding mode control (SMC). This hybrid PLL-SMC approach significantly improves power delivery from photovoltaic (PV) sources, achieving a total harmonic distortion (THD) of 5% and a maximum power point tracking (MPPT) efficiency of 99.1%. Extensive testing demonstrates the inverter's superior performance in grid synchronization, efficiency, and power quality compared to conventional inverters. The results underscore the critical role of advanced GaN-based inverters in enhancing solar energy utilization and advancing renewable energy integration into AC grids. This work sets a new benchmark for PV system integration, contributing to the broader adoption of renewable energy technologies.
Volume: 15
Issue: 3
Page: 2583-2598
Publish at: 2025-06-01

Development of a digital-based fiber tensile testing apparatus to enhance fiber testing accuracy

10.11591/ijaas.v14.i2.pp552-561
Muhammad Iswar , Muhammad Arsyad Suyuti , Rusdi Nur , Ahmad Nurul Muttaqin
Natural fibers are increasingly used in various industries due to their eco-friendly properties and cost-effectiveness. However, current methods for testing the mechanical properties of these materials, such as tensile strength, often face limitations in accuracy and efficiency. This study aims to develop an innovative digital-based fiber tensile testing apparatus to enhance the precision of tensile testing. The research involves the design and construction of the apparatus, utilizing components such as ST37 steel, stepper motors, and Arduino technology. The apparatus was tested using two types of natural fibers, Cocos nucifera L. (coconut fiber) and Sansevieria, to assess their tensile properties. The results showed that although Sansevieria fibers have a smaller diameter, they exhibited higher tensile stress compared to coconut fibers. The developed digital testing apparatus enables more accurate and efficient fiber testing, contributing to the development of stronger and more sustainable materials for industrial applications. The findings of this study highlight the potential of advanced testing equipment in supporting the use of natural fibers in manufacturing and environmental sustainability.
Volume: 14
Issue: 2
Page: 552-561
Publish at: 2025-06-01

Robust deep learning approach for accurate detection of brain tumor and analysis

10.11591/ijece.v15i3.pp3226-3237
Lanke Pallavi , Thati Ramya , Singupurapu Sai Charan , Sirigadha Amith , Thodupunuri Akshay Kumar
Usually, one of the foremost predominant and intricate therapeutic conditions. As broadly perceived, brain tumors are among the foremost significantly harmful circumstances that can radically abbreviate a person’s life expectancy. Various methods are lacking for observing the assortment of tumor sizes, shapes, and areas. When merged with strategies of profound learning, generative adversarial networks (GANs) are competent of catching the measurements, areas, and structures of tumors. Profound learning frameworks will move forward upon the shortage of datasets. It can moreover progress photographs with determination. Classifying and partitioning brain tumors productively is significant. GANs are used in conjunction with an overarching learning handle. A profound learning design called NeuroNet19, could be an intercross of visual geometry group (VGG19) and inverted pyramid pooling module (IPPM) which is utilized to recognize brain tumors. It is clear that, NeuroNet19 employments the foremost exact technique in comparison to all models (DenseNet121, MobileNet, ResNet50, VGG16). The exactness examination gave a Cohen Kappa coefficient of 99% and a F1-score of 99.2%
Volume: 15
Issue: 3
Page: 3226-3237
Publish at: 2025-06-01

ApDeC: a rule generator for alzheimer's disease prediction

10.11591/ijai.v14.i3.pp1772-1780
Sonam Vayaliparambil Maju , Gnana Prakasi Oliver Siryapushpam
Artificial intelligence (AI) paved the way and helping hand for the medical practitioners in various aspects and early disease prediction is one among many. Interdisciplinary research studies on the early prediction of diseases are often analyzed based on the accuracy of the prediction model. But how early these diseases can be predicted will not be answered in many of the research studies unless they have a time series data. This work proposes a machine learning model, ApDeC which solves the above-mentioned problem by generating association rules for the early disease prediction of Alzheimer patients. The ApDeC model calculates the probability of occurrence of eleven Alzheimer disease prediction risk factors and identifies the combination of diseases that can lead to Alzheimer disease. The association rules will be generated by considering the observed combination of risk factors. The research introduces an innovative approach that helps in the early prediction of Alzheimer disease from the risk factors/symptoms. The results show the strong correlation of diabetes and blood pressure with Alzheimer disease.
Volume: 14
Issue: 3
Page: 1772-1780
Publish at: 2025-06-01

Deep learning for predicting drug-related problems in diabetes patients

10.11591/ijece.v15i3.pp2998-3009
Fatima M. Smadi , Qasem A. Al-Radaideh
Machine learning and deep learning have made advances in the healthcare domain. In this study, we aim to apply deep learning models to predict the drug-related problems (DRPs) status for diabetes patients. Also, to determine the appropriate model to use for classification using deep learning algorithms or machine learning methods to investigate which one performed better results for tabular data by comparing the achieved deep learning results with the machine learning methods to figure out which one gives better results. To apply the deep learning models, the same criteria that were applied in the previous study have been implemented in this investigation, and the same dataset was used. The results show that the machine learning algorithms especially the random forests for predicting the DRPs status outperform the deep learning models. For classification tasks in healthcare for tabular data, the findings of this study show that machine learning methods are the appropriate model instead of using deep learning to apply classification.
Volume: 15
Issue: 3
Page: 2998-3009
Publish at: 2025-06-01

Enhancing artificial neural network performance for energy efficiency in laboratories through principal component analysis

10.11591/ijaas.v14.i2.pp310-321
Desmira Desmira , Norazhar Abu Bakar , Mustofa Abi Hamid , Muhammad Hakiki , Affero Ismail , Radinal Fadli
This study investigates energy efficiency challenges during laboratory activities. Inefficient energy use in the practicum phase remains a critical issue, prompting the exploration of innovative forecasting models. This research employs artificial neural network (ANN) models integrated with principal component analysis (PCA) to predict energy consumption and optimize usage. The findings reveal that PCA components, including eigenvalues, eigenvectors, and matrix covariance values, significantly influence the ANN model's performance in forecasting energy production. The ANN training achieved a high correlation coefficient (R=1) with a mean squared error (MSE) of 0.045931 after 200,000 epochs, demonstrating the model's robustness. While testing results showed a moderate correlation (R=0.46169), the models demonstrated potential for refinement and scalability. This integration of ANN and PCA models provides a reliable framework for accurately forecasting energy usage, offering an effective strategy to enhance energy efficiency in laboratory settings. By optimizing energy consumption, this approach has the potential to reduce operational costs and environmental impact. The strong performance metrics highlight the practical utility of these models in educational contexts, contributing to sustainable energy management and better resource allocation. Furthermore, the reduction in energy-related environmental impacts underscores the broader applicability of these models for fostering sustainable development in similar contexts.
Volume: 14
Issue: 2
Page: 310-321
Publish at: 2025-06-01

Viral hepatitis morbidity and mortality data in major urban cities in the Philippines

10.11591/ijphs.v14i2.24577
Rael S. Manriquez , Mark Anthony J. Torres , Cesar G. Demayo
This study investigates the transmission, impact, and prevention of viral hepatitis A (HAV), hepatitis B (HBV), hepatitis C (HCV), hepatitis D (HDV), and hepatitis E (HEV) in the National Capital Region (NCR) and Region 7, Philippines, from 1960 to 2020. These infections significantly contribute to liver complications, including cirrhosis and hepatocellular carcinoma, affecting mental well-being and posing risks to pregnant women. Although hepatitis mortality is notable, complete treatment can mitigate the risk. Transmission occurs through various routes, such as blood products, body secretions, and perinatal routes. The study underscores the importance of understanding transmission and implementing screening and prevention measures. Vaccination, particularly for Hepatitis A and B, is crucial, reshaping disease epidemiology through universal infant immunization. Challenges like low vaccination coverage persist, especially among children and healthcare workers. Analyzing mortality data reveals a significant recent decrease attributed to government efforts and vaccination programs since 1995. Despite regional variations, mortality remains relatively low. The study recommends prioritizing and expanding vaccination programs, raising awareness, improving healthcare accessibility, and strengthening surveillance systems. Coupled with community engagement, these measures promise sustained success against viral hepatitis, reinforcing the observed trend in mortality reduction.
Volume: 14
Issue: 2
Page: 1015-1021
Publish at: 2025-06-01

A hybrid convolutional neural network-recurrent neural network approach for breast cancer detection through Mask R-CNN and ARI-TFMOA optimization

10.11591/ijece.v15i3.pp3084-3094
Keshetti Sreekala , Srilatha Yalamati , Annemneedi Lakshmanarao , Gubbala Kumari , Tanapaneni Muni Kumari , Venkata Subbaiah Desanamukula
This paper presents a novel hybrid deep learning-based approach for breast cancer detection, addressing critical challenges such as overfitting and performance degradation in varying data conditions. Unlike traditional methods that struggle with detection accuracy, this work integrates a unique combination of advanced segmentation and classification techniques. The segmentation phase leverages Mask region-based convolutional neural network (R-CNN), enhanced by the adaptive random increment-based tomtit flock metaheuristic optimization algorithm (ARI-TFMOA), a novel algorithm inspired by natural flocking behavior. ARI-TFMOA fine-tunes Mask R-CNN parameters, achieving improved feature extraction and segmentation precision while ensuring adaptability to diverse datasets. For classification, a hybrid convolutional neural network-recurrent neural network (CNN-RNN) model is introduced, combining spatial feature extraction by CNNs with temporal pattern recognition by RNNs, resulting in a more nuanced and comprehensive analysis of breast cancer images. The proposed framework achieved significant advancements over existing methods, demonstrating improved performance. This hybrid integration of ARI-TFMOA and Hybrid CNN-RNN models represents a unique contribution, enabling robust, accurate, and efficient breast cancer detection.
Volume: 15
Issue: 3
Page: 3084-3094
Publish at: 2025-06-01

Leukemia detection using SegNet and faster region-based convolutional neural network

10.11591/ijece.v15i3.pp3028-3038
Della Reasa Valiaveetil , T. Kanimozhi
Prevention of cancer is mostly attained by surveillance of the transformation zones. White blood cells (WBCs) are established in the bone marrow and intemperate growth of WBC leads to leukemia. Hematologists examine the microscopic images in manual method for predicting leukemia, but it is very complex process and without any guaranteed for accurate. In this proposed study, deep learning techniques involved to segment and classify the three types of leukemia like acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML) and chronic lymphocytic leukemia (CLL) using the BioGps dataset. The purpose of deep learning in medical science enhances the accuracy and precision of determining leukemia in early stages. In this study, introducing a sigmoid stretching (SS) in pixel enhancement for preprocessing; SegNet (St) is comfort to extract the structural features of the leukocytes and to segment the normal and blast cells for a clear classification; faster region-based convolutional neural network (faster R- CNN) carried under the process of classification and optimization done by dragon fly algorithm. The result of this work achieves best accuracy related to the existing techniques of convolutional neural network (CNN) such as support vector machine (SVM), k-nearest neighbors (kNN) and Bayesian model. This study achieves the accuracy rate of 97%, precision rate of 94% and sensitivity rate of 90% respectively with low complexity.
Volume: 15
Issue: 3
Page: 3028-3038
Publish at: 2025-06-01

Genetic algorithm-adapted activation function optimization of deep learning framework for breast mass cancer classification in mammogram images

10.11591/ijece.v15i3.pp2820-2833
Noor Fadzilah Razali , Iza Sazanita Isa , Siti Noraini Sulaiman , Muhammad Khusairi Osman , Noor Khairiah A. Karim , Dayang Suhaida Awang Damit
The convolutional neural network (CNN) has been explored for mammogram cancer classification to aid radiologists. CNNs require multiple convolution and non-linearity repetitions to learn data sparsity, but deeper networks often face the vanishing gradient effect, which hinders effective learning. The rectified linear unit (ReLU) activation function activates neurons only when the output exceeds zero, limiting activation and potentially lowering performance. This study proposes an adaptive ReLU based on a genetic algorithm (GA) to determine the optimal threshold for neuron activation, thus improving the restrictive nature of the original ReLU. We compared performances on the INbreast and IPPT-mammo mammogram datasets using ReLU and leakyReLU activation functions. Results show accuracy improvements from 95.0% to 97.01% for INbreast and 84.9% to 87.4% for IPPT-mammo with ReLU and from 93.03% to 99.0% for INbreast and 84.03% to 91.06% for IPPT-mammo with leakyReLU. Significant accuracy improvements were observed for breast cancer classification in mammograms, demonstrating its potential to aid radiologists with more robust and reliable diagnostic tools.
Volume: 15
Issue: 3
Page: 2820-2833
Publish at: 2025-06-01

Impact of natural-white and red-blue light-emitting diode lighting on hydroponic basil growth and energy efficiency

10.11591/ijaas.v14.i2.pp406-415
Chaiyant Boonmee , Warunee Srisongkram , Wipada Wongsuriya , Patcharanan Sritanauthaikorn , Paiboon Kiatsookkanatorn , Napat Watjanatepin
Advanced phosphor-converted white light-emitting diodes (pc-WLEDs) have been developed to mimic the natural sunlight spectrum, potentially enhancing plant growth compared to traditional red-blue (R-B) LEDs. This study aimed to compare the effects of natural-white pc-WLED (nsW-pcLED) and conventional R-B LED (R:B 3.24) on the growth, yield, and energy efficiency of hydroponically grown sweet basil. It was cultivated in a deep-water culture system under identical conditions with a photosynthetic photon flux density (PPFD) of 200±10 µmol·m⁻²·s⁻¹ and a 16/8 light/dark photoperiod over 28 days. Key growth parameters, including plant height, stem diameter, leaf number, and plant fresh weight (PFW), were measured, while energy consumption was recorded to assess efficiency. Results indicated that nsW-pcLED significantly enhanced growth, with plants achieving an average height of 44.30±1.51 cm, stem diameter of 6.68±0.21 mm, and a PFW of 34.20±6.12 g, compared to 35.88±4.05 cm, 4.66±0.88 mm, and 23.02±5.26 g under R-B LED (p <0.05), respectively. The nsW-pcLED treatment produced an average net growth of 1,221 g·m⁻² versus 536.43 g·m⁻² for R-B LED and delivered 33.05 g·m⁻²·kW·h⁻¹ compared to 11.17 g·m⁻²·kW·h⁻¹, while consuming 23% less energy. These findings highlight nsW-pcLED’s superior performance for indoor hydroponic cultivation. Future studies should explore its application in large-scale systems and across diverse crop species.
Volume: 14
Issue: 2
Page: 406-415
Publish at: 2025-06-01

Highly sensitive microwave sensor for metallic mine detection

10.11591/ijece.v15i3.pp2631-2641
Maged A. Aldhaeebi , Thamer S. Almoneef
This study introduces an innovative microwave system for detecting buried metallic landmines, providing an alternative to conventional imaging approaches. The system consists of two highly sensitive sensors, each configured with identical antennas arranged in a triangular formation to enhance sensitivity. The proposed microwave sensors exhibit exceptional sensitivity in detecting metallic landmines buried at various depths within sand and at different distances. Simulation and experimental studies were conducted using a foam box filled with sand and a metallic cube to simulate a landmine. The sensor’s sensitivity is evidenced by shifts in both the magnitude and phase of insertion loss (𝑆21) between scenarios with and without a metallic mine, attributed to differences in dielectric properties between the sand and the mine in the microwave spectrum. The results from both simulations and experiments confirm the sensor’s capability to detect metallic mines at varying depths within the sand medium. The proposed system offers significant advantages over imaging technologies for mine detection, including cost-effectiveness, simplicity, and ease of data processing without the need for complex imaging algorithms.
Volume: 15
Issue: 3
Page: 2631-2641
Publish at: 2025-06-01

Carbonized mangrove wood as photothermal material for solar water desalination

10.11591/ijaas.v14.i2.pp542-551
Dolfie Paulus Pandara , Kristina Unso , Maria Daurina Bobanto , Gerald Hendrik Tamuntuan , Ping Astony Angmalisang , Ferdy Ferdy , Vistarani Arini Tiwow , Maureen Kumaunang
The investigation into the physical properties of carbonized mangrove wood (CMW) is essential for its development as an efficient solar heat absorber. This study explores the physical characteristics of CMW and its potential application in solar desalination. Initially, the mangrove wood was cleaned with running water, followed by ultrasonication at a frequency of 42 kHz in 96% ethanol for 5 minutes, and then heated at 125 °C for 2 hours. The carbonization process was conducted in a furnace for 1 hour at temperatures of 400, 500, and 600 °C. The physical properties of CMW were analyzed using an X-ray diffractometer (XRD), Fourier transform infrared spectroscopy (FTIR), energy dispersive spectroscopy, and scanning electron microscopy (SEM). The findings revealed the formation of a carbon structure at 2 theta angles of approximately 24.08, 23.26, and 23.16°, with carbon contents of 45.05, 36.86, and 39.37%, respectively. CMW was identified as a porous material, making it highly effective for sunlight absorption in seawater evaporation. The hydroxyl content within the CMW structure enhanced its water evaporation capabilities. In experimental investigations aimed at desalinating seawater, a 300-watt halogen lamp was positioned 15 centimeters above the CMW's surface, resulting in an evaporation rate of 5.33 kg.m-2.h-1. CMW shows significant promise as a solar evaporator.
Volume: 14
Issue: 2
Page: 542-551
Publish at: 2025-06-01

Harnessing speed breakers potentials for electricity generation: a case study of Covenant University

10.11591/ijece.v15i3.pp2669-2680
Hope Evwieroghene Orovwode , John Amanesi Abubakar , Olutunde Oluwatimileyin Josiah , Ademola Abdullkareem
The global imperative to transition towards sustainable energy sources has sparked innovative solutions for energy generation and environmental conservation challenges. As fossil fuel usage for power generation continues to raise environmental concerns, converting kinetic energy from vehicular motion via speed breakers presents a unique avenue for renewable power production. This study explores the concept of utilizing speed breakers as a means of electricity generation to power little power-consuming but critical load, with Covenant University serving as a pertinent case study. This research investigates the technical, economic, and environmental implications of implementing speed breaker-based electricity generation within Covenant University. Analyzing the university's energy consumption patterns showed that some loads do not require much power but are critical. Street lighting is one of such loads. This study discerns the potential contribution of speed breaker-generated electricity to address energy demands by simulation and constructing a prototype. Advanced engineering tools, such as simulation software Fusion 360 and Proteus 8.0, were employed to model and integrate the roller speed breaker mechanism with the electrical infrastructure. The findings offer valuable insights into the viability of speed breaker-generated electricity as an alternative energy source, paving the way for sustainable energy practices in educational institutions and beyond.
Volume: 15
Issue: 3
Page: 2669-2680
Publish at: 2025-06-01
1 2 3
...
1995 1996
Show 185 of 1996

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