Asian Journal of Research in Computer Science
Not a member yet
792 research outputs found
Sort by
Optimizing DevOps for Critical Systems
No full textThis article examines the impact of DevOps practices on improving the productivity of software development teams and managing quality variability in the maintenance of critical systems. Based on an analysis of the MONTE system developed by the Mission Design and Navigation Software group at the Jet Propulsion Laboratory (JPL), a comparative assessment was conducted of key metrics, including defect density, release frequency, and incident response time, before and after the implementation of DevOps. The application of time-series analysis methods demonstrated that systematic integration of automation, continuous integration and delivery, and infrastructure as code contributes to reducing operational risks and enhancing process efficiency. DevOps cycles in mission-critical systems function as continuously iterative pipelines for deployment, testing, and delivery. They are built on principles of closed-loop feedback and adaptive change management, where each iteration includes formal static verification and automated deployment with compliance checks against regulatory standards. DevOps methodologies in such systems are grounded in Infrastructure as Code and GitOps practices, combined with advanced release strategies and blue/green deployment models. These approaches support a high level of observability, predictive monitoring based on reliability metrics, and continuous risk management through integrated security controls and configuration management components. Additionally, the article provides an overview of strategies and recommendations for the successful adoption of DevOps, considering organizational and technological prerequisites, changes in corporate culture, and the need for skill development among specialists. The study results confirm the hypothesis that DevOps approaches serve as an effective tool for enhancing stability, reliability, and team productivity in the operation of mission-critical systems. The information presented will be valuable for researchers and IT professionals, as well as for managers seeking to integrate advanced DevOps methodologies to improve team performance and optimize business processes
A Novel Method for Information Hiding in Images
No full textInformation security is a fundamental principle during data transmission. The main threat to information security is hacking. They are focusing on finding new techniques to breach encryption methods. Therefore, developing new approaches for hiding information against hacker attacks is a potential solution. Different steganography methods are available for hiding information, such as in texts, disk space, network packets, audio and images. In contrast to cryptography, which is concerned with data encryption. Steganography is the art of hiding transmitted information by using seemingly harmless carriers to hide the presence of information. Steganography is cost-effective and unrelated to hardware components. This study aims to develop a novel method for hiding information in images at low cost. The proposed method is combing between mathematical transformations (i.e., DCT transformation) and image characteristics (i.e., image components, and location of the image’s pixel). The JPEG colour images were used to apply the proposed method as a cover image. The text message was spread on the cover image without affecting its size. The proposed method has shown significant results in information hiding with a PSNR of 60.65 dB that exceeds the typical threshold of PSNR, which is near 40 dB to 60 dB. It can be utilised in data communication, user authentication, finance, and banking operations
Deep Learning for Medical Imaging: A Comprehensive Review of NLP Algorithms, Advancements and Challenges
No full textThe present study examines the impact of Artificial Intelligence (AI) and Machine Learning (ML) on medical imaging, focusing on the demand for automated analysis of unstructured data in Electronic Medical Records (EMRs). It addresses the challenges in extracting knowledge from this data despite advancements in Natural Language Processing (NLP) and image processing. This review covers key principles and challenges of applying ML, including the use of algorithms like logistic regression, decision trees, and neural networks to classify and predict illnesses. It also explores various machine learning techniques such as supervised, unsupervised, and reinforcement learning, and emphasizes the importance of data preprocessing, feature selection, and model evaluation. The review also highlights various AI applications in medical imaging, including image segmentation, classification, registration, and reconstruction across modalities like X-ray, CT, MRI, and ultrasound. It also points out AI’s potential in enhancing robotic surgery through innovative techniques such as holography and attention models for early disease detection. While deep learning shows promise in disease diagnosis, the lack of large, annotated datasets remains a barrier. The authors note the progress in unsupervised and semi-supervised learning methods to tackle this issue. They stress the need for collaboration between healthcare professionals and AI experts to improve the interpretability of deep learning models. Ultimately, the review concludes that while AI has the potential to improve diagnostic accuracy and treatment strategies, challenges like data availability and ethical considerations must be addressed for successful implementation in healthcare
Artificial Intelligence Models for Predicting Fertility Transitions: A Systematic Review
No full textThe demographic transition from high to low fertility has profound societal impacts. Recent years have seen growing interest in applying artificial intelligence (AI) to forecast fertility outcomes at both individual and population levels. This review systematically surveys empirical studies (2010–2025) using AI models to predict fertility-related transitions.
I screened peer-reviewed articles and identified 30 studies addressing fertility forecasts or outcomes. Most studies apply machine learning to demographic surveys, vital statistics, or personal health data. I report model types, input features, training data, and evaluation metrics with concrete examples.
Tadese et al. (2024) used Random Forest and XGBoost on Nigeria Demographic and Health Survey data (n=37,581) to predict women\u27s fertility preferences, achieving accuracy of 0.92 and AUC of 0.98. Tzitiridou-Chatzopoulou et al. (2024) applied XGBoost to forecast monthly birth counts in Scotland, obtaining MAE of 0.32 (vs 0.44 for ARIMA). Deep models have also shown promise. Xue et al. (2024) developed an attention-augmented LSTM (BRP-Net) using economic and demographic features to predict regional birth rates with RMSE of 0.22.
I also summarize fairness and equity analyses. Few studies explicitly audit demographic biases. One preeclampsia risk model reported a correctable bias against Black women. Most models neglect performance evaluations for underrepresented groups.
I examine regulatory readiness, clinical adoption challenges, and the need for interpretability and user-centered design. Overall, AI shows promise for fertility forecasting but requires careful validation, fairness auditing, and stakeholder engagement before clinical or policy deployment
AI and IoT Integration for Predictive Maintenance and Risk Management in Smart Manufacturing
No full textPredictive maintenance has emerged as a cornerstone of Industry 4.0, enabling manufacturers to proactively identify and address equipment failures, minimize unplanned downtime, and optimize operational costs. However, realizing effective predictive maintenance in smart manufacturing environments requires overcoming persistent challenges related to real-time data communication, cybersecurity vulnerabilities, and system scalability. This study addresses these gaps by investigating the integration of Artificial Intelligence (AI) and Internet of Things (IoT) technologies for predictive maintenance using the NASA C-MAPSS dataset. A quantitative methodology was employed, including transmission protocol analysis, cybersecurity assessment with an Isolation Forest-based Intrusion Detection System, scalability evaluation on edge and cloud infrastructures, and predictive modeling with Long Short-Term Memory (LSTM) networks. This research develops and empirically validates an integrated AI-IoT framework that unifies communication efficiency, cybersecurity resilience, and predictive modeling, representing a novel contribution to the state of the art. Results show MQTT achieved the lowest latency (50.21 ms), the IDS attained a Precision of 92.31%, edge systems supported up to 3352 MB before degradation, and the LSTM model outperformed linear regression with an RMSE of 14.25 and R² of 0.92. The study recommends that manufacturers adopt MQTT for efficient real-time communication, deploy AI-driven intrusion detection to safeguard predictive analytics, invest in scalable edge computing infrastructures, and implement deep learning models within hybrid edge-cloud architectures to enhance predictive maintenance reliability and support immediate, practical deployment in Industry 4.0 manufacturing systems
Leveraging AI-Powered Conversational Agents to Mitigate Vaccine Hesitancy in Low-Resource African Contexts: A Public Health Framework
No full textVaccine hesitancy remains a major public health concern across Africa, driven by misinformation, cultural beliefs, and limited access to accurate health communication. With growing mobile and internet access, AI-powered conversational agents (chatbots) offer a promising means of improving vaccine literacy and trust in low-resource settings such as Nigeria. A qualitative review of studies from PubMed, Scopus, and IEEE Xplore, along with WHO and Africa CDC reports, was conducted to examine chatbot applications in healthcare. Findings informed the design of a multimodal framework that integrates text, voice, and visuals in indigenous languages (Igbo, Yoruba, Hausa) for inclusive communication. A pilot design covering urban (Lagos) and rural (Abia) populations was proposed to evaluate comprehension, engagement, and accessibility. The review shows that culturally localized chatbots can substantially enhance vaccine literacy. Projected outcomes indicate up to a 70% improvement in comprehension and a 60% increase in engagement when multimodal features and linguistic adaptation are incorporated. Ethical and infrastructural considerations remain key for sustainable deployment. AI-driven conversational agents provide a scalable, low-cost solution to vaccine hesitancy in Africa. By aligning technology with cultural and linguistic diversity, they can bridge communication gaps, counter misinformation, and strengthen public health awareness. This study contributes a context-driven framework for integrating AI into community-based vaccine education
A Real-time Multi-lingual Android Translator Integrating Text and Voice Recognition
No full textIn an increasingly interconnected global environment, the need for seamless multilingual communication via mobile devices is growing. This paper presents the design, development, and evaluation of a real-time multi-lingual Android system application translator (RT-MLAST), which integrates both text and voice recognition capabilities to convert spoken or typed input in one language to spoken and/or written output in a target language. The system pipeline comprises automatic speech recognition (ASR), neural machine translation (NMT), and text-to-speech (TTS) modules, augmented by language detection and multilingual support features. The study adopted a design science research (DSR) approach, which focuses on the creation and evaluation of an innovative artefact to solve a real-world problem. The ASR component was implemented using TensorFlow Lite to ensure compatibility with Android mobile devices. To enable efficient on-device operation, the NMT model was compressed using quantisation and pruning techniques. In order to collect data, a questionnaire was set up to check users\u27 perception of the designed application. The Android application was implemented following the Model-View-Controller (MVC) design pattern to ensure modularity and scalability. The contributions of this work include a mobile-device real-time translator architecture, empirical measurements in the Android domain, and insights for deploying multilingual systems on resource-constrained devices. The results showed that the application achieved a whopping result of 99% accuracy when it was used for translation from one language to another using the intended means (Voice, text or image). The overall user acceptability based on the aspects was 3.50 out of 5. This suggested that users generally find the app quite acceptable, with only a few areas that could be improved (e.g., translation accuracy and responsiveness). For the Yoruba language, the application exhibited high accuracy in translating simple and direct sentences. However, it faced challenges when handling complex expressions and idiomatic phrases unique to Yoruba culture, which require a deeper understanding of linguistic nuances. The results show that the proposed system achieved competitive translation accuracy and high usability. The discussion implications for global communication, tourism, and education, and proposes directions for future work
Novel Implementation of Vehicle Drivers Alcohol and Temperature Monitoring System
No full textDrunk driving is one of the major cause of road accidents especially in Nigeria where there is no controls or checks to the level of alcohol content of drivers. In this study, a vehicle alcohol and temperature monitoring system was simulated and constructed using an Arduino Uno microcontroller ATMega328. The circuit was simulated in Proteus8.1 design suit comprising an alcohol sensor, contactless temperature sensor, motor/relay driver, ignition, 16×2 liquid crystal display, and 5V DC power. The simulated circuit was constructed on a printed circuit board and tested for continuity and power ‘ON’. An evaluation performance test was also carried out using 40 volunteers that have consumed varying amounts of alcohol content ranging from two to fifteen standard drink and their blood alcohol content as well as body temperature were measured. The true positive, false positive, true negative and false negative were recorded. Results showed that the constructed device has sensitivity of 92% and specificity of 75% depicting that the system can correctly identify 92% presence of alcohol and 75% of alcohol absence with 90% accuracy. It however, failed to identify only 8% alcohol presence and 25% alcohol absence. The device if installed in a vehicle could be significantly useful in averting road accidents caused by drunk driving. Ensuring the public and commercial drivers have a clear understanding of the law and raising awareness of the impact drink drive has on road users is important to reducing its prevalence
Preliminary Differential Diagnosis of Pneumonia Disease: An Innovative Approach Using an Expert System Based on Rules
No full textThis study focuses on the development of a rule-based expert system for diagnosing people with pneumonic infections. Pneumonia is the most common respiratory disease causing death worldwide, and its diagnosis is difficult due to clinical symptoms similar to other respiratory diseases. As a result, doctors often order multiple tests before making a decision, leading to high costs and longer wait times. The expert system developed in this study aims to help doctors and patients distinguish between pneumonia and other diseases such as cancer, chronic bronchitis and tuberculosis. The system takes symptoms such as fever, lack of appetite, cough, chills, hemoptysis and chest pain as input and produces pneumonia as output. The system has gone through four stages of development: definition of a knowledge system, design, implementation, evaluation and testing. The study is based on a dataset made up of 152 medical records including patients with respiratory symptoms similar to those of pneumonia. This data comes from hospital sources or medical databases, integrating information on medical history, chest imaging results and biological analyses. Validation of the system was carried out by comparing its performance to diagnoses made by specialists. The results indicate a diagnostic accuracy of 76%, demonstrating the effectiveness of the system in differentiating pneumonia from other respiratory conditions such as bronchitis, tuberculosis or pulmonary embolism. The study concludes that this rule-based expert system provides a promising tool to assist clinicians in the differential diagnosis of pneumonia, particularly in resource-limited settings where specialized medical expertise may be lacking. The Cohen\u27s Kappa coefficient (κ\kappaκ) is approximately 0.76, indicating a substantial agreement between the expert system and the doctors. This suggests that the expert system performs well but still has room for improvement
Advancing Security in Cloud-based Patient Information Systems with Quantum-resistant Encryption for Healthcare Data
No full textThe increasing adoption of cloud-based patient information systems in healthcare enhances efficiency and accessibility but introduces significant cybersecurity risks. Traditional encryption methods, such as RSA and ECC, are becoming vulnerable due to rapid advancements in quantum computing, necessitating the transition to quantum-resistant encryption. This study evaluates the effectiveness of post-quantum cryptographic (PQC) solutions in securing cloud-stored healthcare data. Using publicly available datasets, including HHS breach records, NIST PQC benchmarks, and HIMSS cybersecurity reports, this research identifies trends in healthcare data breaches, encryption adoption, and regulatory penalties. Findings indicate a strong correlation between increased cloud adoption and cybersecurity breaches, highlighting the urgent need for enhanced encryption. Post-quantum cryptographic algorithms, particularly CRYSTALS-Kyber and CRYSTALS-Dilithium, outperform traditional encryption methods in terms of security and computational efficiency, making them viable for healthcare applications. Institutions that have adopted PQC show a marked decline in regulatory fines, reinforcing its role in compliance and risk mitigation. This study underscores the need for early PQC adoption, increased investment in cybersecurity training, and regulatory collaboration to ensure seamless integration of quantum-resistant encryption. The research provides critical insights into the evolving cybersecurity landscape, emphasizing the necessity of transitioning to PQC to safeguard patient data against future quantum threats. These findings serve as a strategic guide for healthcare institutions and policymakers, advocating for proactive encryption strategies that align with regulatory standards and ensure long-term data security