Ifedayo A. Olusola, Olusegun G. Lala, Adebayo Aroyehun et al. (4 contributing authors)
Coronary Heart Disease (CHD) remains one of the leading causes of global morbidity and mortality, demanding innovative approaches for effective clinical management and medication administration. This study presents an intelligent model designed to support drug administration in patients with CHD using Artificial Neural Networks (ANN), structured clinical workflows, and stakeholder-informed system requirements. Data for model development were obtained using structured interviews with cardiologists, pharmacists, and administrative personnel from UNIOSUN Teaching Hospital, Osogbo, Osun State and LAUTECH Teaching Hospital, Ogbomoso, Oyo, State. In addition, a merged heart disease dataset from Kaggle containing 1,190 instances and 11 clinically relevant features. Preprocessing procedures including data cleaning, feature selection, normalization, and categorical encoding were performed on the acquired data. The proposed system was designed using UML tools to model workflows, class interactions, and user roles, ensuring alignment with real-world clinical processes. The ANN model was developed and integrated into a functional clinical information system using Node.js, Flutter, and RESTful API communication. Performance evaluation revealed that the implemented model demonstrated strong potential for improving drug administration accuracy, enhancing clinical decision-making, and streamlining pharmaceutical operations. The integrated MedAd4CHD platform further ensures secure data flow, real-time patient information access, and end-to-end medication lifecycle management. This study provides a practical, scalable framework for intelligent CHD medication administration and establishes a foundation for future clinical validation.
Coronary Heart DiseaseIntelligent Drug AdministrationClinical Decision Support SystemHealthcare AutomationMachine Learning in Medicine
Tai Solarin University of Education (TASUED) has faced challenges in the efficiency and coordination of its clinic emergency response system, leading to delays and raising concerns about the safety of students, staff, and visitors. This study aims to design and implement a computerized TASUED Clinic Emergency Response System to enhance emergency management capabilities. Leveraging technology, the proposed system seeks to streamline communication channels, improve response times, and ensure effective coordination during emergencies. Observational methods and interviews conducted at the TASUED Medical Center revealed inefficiencies in patient registration, waiting times, and staff scheduling. Based on these findings, the system was developed using a top-down design approach and modular testing strategy, ensuring robust and reliable functionality. The implementation includes a user-friendly login interface and a comprehensive dashboard for healthcare professionals, facilitating efficient emergency response. This system's deployment is a significant step towards improving emergency management at TASUED and can serve as a model for other institutions facing similar challenges.
There is limited availability and usage of cooking gas leakage detectors for fire safety purposes in developing countries of Africa owing to their high cost, unstable electric power supply source and inefficient gas sensor architecture although over 80% of the population relies on liquefied petroleum gas (LPG) for cooking. Unfortunately, LPG cookers are associated with most fire and gas explosions in this region because of its high flammability emanating from gas leakages within the system, leading to environmental hazards. This paper presents the development of low cost, reduced electric power consumption, miniaturized LPG gas leakage detector. The developed device implements unique system for the detection of LPG gas, triggers an alarm through a buzzer upon detection of leakage, and delivers a warning short message service (SMS) to the enrolled users prompting them of the LPG gas leakage or incomplete combustion of the LPG gas at the burners. The deployed circuitry demonstrates high levels of efficiency, reliability, and is affordable, therefore is attractive for low-income countries of Africa.
Nigerian correctional facilities rely on inefficient manual record-keeping processes, hindering effective prisoner information management and inter-agency coordination. This study aims to develop a comprehensive, centralized, web-based Prison Management System (PMS) to modernize and streamline operations in Nigerian correctional centers. The research employed a multifaceted data collection approach, including interviews, observations, and form evaluations. System development utilized the Incremental Model, incorporating MySQL, PHP, and JavaWeb technologies. The proposed PMS introduces a unified platform consolidating prisoner profiles, including demographic details, biometric data, parole status, visitor records, and departure logs. The system enables remote access, real-time information exchange, and intelligent administration through cloud computing integration. This digital transformation promises to elevate the Nigerian correctional system's efficiency, accountability, and inter-agency coordination, contributing to a more effective and secure justice framework.
Oguntunde, B. O., Ologunye et al. (7 contributing authors)
The mandatory wearing of face masks orchestrated by the COVID-19 pandemic has brought some complexities to the ability of face recognition systems in identifying such faces. Thus, this study examined the performance of some selected machine learning algorithms: Convolutional Neural Network (CNN), Linear discriminant analysis (LDA), Logistic Regression (LR), K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Naïve Bayes (NB) and Decision Tree (DT) for the recognition of masked and unmasked human faces; and developed a face recognition system based on the algorithm with the best performance. The dataset was composed of 11,792 masked and unmasked facial images collected from Kaggle online repository. The set was divided into 70%, 20% and 10% respectively for training, testing and validation each for the masked and the unmasked face images. The images were augmented randomly for training, with some rotated 30 degree, some zoomed 20%, some shifted horizontally by 10% in width, some shifted vertically by 10% in height and some flipped horizontally. The performance analysis of the algorithms presented accuracies of 99% and 99% for the CNN; 93% and 98% for LDA; 93% and 93% for LR; 86% and 78% for the NB; 70% and 70% for the KNN; 64% and 45% for the DT; and 92% and 98% for the SVM respectively for training and cross validation. Overall, the CNN recorded the highest recognition accuracy. Thus, a CNN user-friendly face recognition, system was developed, and tested with a number of real-life human masked and unmasked faces which showed excellent recognition performance. The implementation was carried out with appropriate Python machine libraries. The developed system could be very useful for access control and surveillance. Therefore, this study recommends the adoption of the developed systems for face recognition-based security system.
This paper presents an implementation of a home automation system using smart phone Bluetooth with Arduino for the switching and control of home appliances. It aimed at reducing stress for the physically challenged, elderly, and younger children with an easy control of the home appliances and likewise reduction of risk of electrocution from the source of control. With this automated control unit, users can comfortably operate their electrical installations or appliances from distances ranging from 30 m to 100 m away from the point of control using smart phone. The control unit comprises a four-channel relay module, an Arduino Uno board (AUB) based on the ATmega328 microcontroller, an HC-05 Bluetooth module, and a manual backup switch or alternative. The central processing unit is the Arduino Uno board to which every other component is connected. The Load (devices/appliances) to be controlled or operated is connected to the Arduino Uno board (AUB) via the relay. The Arduino board is programmed using the Arduino integrated development environment (IDE), with instructions that controls other units of the circuits. An Arduino Bluetooth controller is installed on the phone, to be used as a remote controller and to enable communication interface between the phone and the Arduino based control unit. In other words, the Bluetooth module connected to the Arduino board receives signals and/or instructions through phone Bluetooth via controller installed on the phone. An LCD (led crystal display) is used as display output of information uploaded on the Arduino Uno board to know the working state of the circuit. It was found that the automated control unit is of great benefit to users in terms of convenience in operating their appliances, including the elderly, and physically challenged people as it reduces their stress and difficulty in moving from point to point in order to control their home lightings and appliances.
Balogun, Oluwatimilehin B.1, and Osazuwa Isaac B.2
Physical geodetic survey of the Nigeria landmass was undertaken with the aim of furnishing engineers and scientists with vital geodetic information which have not been adequately available but required for the development of a standard national economic framework. The basic data used for the study were the absolute gravity field values, and the recorded coordinates and elevations at the Primary Gravity Network of Nigeria (PGNN) base stations. The absolute gravity data were processed using the appropriate mathematical and computational procedures to deduce geodetic information which include geoid undulation, height above the geoid, deflections of the vertical, distance from the earth’s centre of mass to the reference ellipsoid and geoid, and the gravity potential value at the geoid. Comparison was also made between the WGS’84 and the geoid in the study area. Geoid undulations were found to range from 16.53 to 28.35 m and the deflections of the vertical ranged between 0.738685 to 10.28328 arc second. Height above the topography ranged from -16.38 m in Warri to +1257.13 in Jos. The distance from the earth’s centre of mass to the WGS’84 reference ellipsoidal surface was maximum at Oron (6377988.04 m) and least at Illela (6376941.31 m). Similarly, the distance from the earth’s centre of mass to the geoid was also maximum at Oron (6378008.39 m) and minimum in Illela (6376961.659 m). The interpolation of the distances from the earth’s centre of mass to the spheroidal surface at each base station gave the definition of the WGS’84 ellipsoid over the continental Nigeria landmass while the interpolation of the distances from the earth’s centre of mass to the geoidal surface gave the definition of the geoid over the continental Nigeria landmass. The average gravity potential at the geoid was evaluated as 62375050.41 m 2 s -2 . The geoid over the continental Nigeria landmass was found to be comparably smooth as the WGS’84 ellipsoid over the region though they differ in terms of vertical separation. The quantitative difference in mGal corresponding to the vertical separation between the WGS’84 reference ellipsoid and the geoid was computed as 6.8042 mGal. The values of the deflections of the vertical at each station of the PGNN gave the angular correction to be applied when combining satellite-based ellipsoidal measurements and geoid-based physical geodetic measurement. Based on the significant range of the deflections of the vertical across the continental area of Nigeria, it was concluded that the physical structure of the lithosphere beneath the study area must be highly complex and heterogeneous.
Physical geodetic survey Primary Gravity Network of Nigeria (PGNN)geoid undulationsdeflections of the vertical
F. A. Kasali, J. A. Balogun, O. D. Adekola et al. (4 contributing authors)
In the education sector, constructing a personalized timetable is an important aspect that should be given uppermost priorities when it comes to academic planning and implementation. Basically, in most schools, the timetable is done to encompass all courses, segregation only comes in departmentally, by faculty or by programs being studied by the students involved. This could be cumbersome especially for schools with numerous programs, departments or faculties as students and faculty members have to glance through such complex timetables before they are able to map out theirs. This could lead to resource wastage in terms of time, misinterpretation by students involved, clashing of courses being taken by students, poor utilization of available venues amongst other issues. In other to proffer solutions to some of the identified problems, a personalized and user-centric timetable system was designed for the College of Basic and Applied Sciences (CBAS), Mountain Top University, Ogun State, Nigeria. System design was specified using unified modeling language (UML) diagrams such as use-case, sequence and activity diagrams. The system was implemented using mobile application development technologies. The android based timetable system presented an administrative module who could schedule lecture periods, specify time amongst other constraints and overall management of course scheduling based on courses registered for by the students. Evaluation of the system was done using Post Study System Usability Questionnaire (PSSUQ) version 3 measuring tool and analysed using descriptive statistics with SPSS version 21. One hundred and twenty five (125) participants were chosen randomly and purposefully for the study. Results of analysis gave the perceived satisfaction score as 2.56, system usefulness with 1.98, information quality with a mean score of 2.72 while interface quality performed the lowest with a mean score of 3.19 since in PSSUQ analysis, the higher the mean score, the poorer the performance. The timetable system was basically designed to reduce the problem of accessibility and better convenience especially on the part of the students. Future works would also consider a fully automated system with the use of scheduling algorithms and the integration of examination and invigilation scheduling in the same system.
Uzoma Ifeanyi Oduah,\r\nOnyedikachi Alanwoko and Shola Folarin
We hereby present an advanced road vehicle safety device. The device tackles the problem of loss of lives, arising from late medical attention to care for accident victims after an accident as a solution that enables fast response to persons involved in ghastly vehicle accidents. It uses ultrasonic sensor to detect the changes in the body of the vehicle due to a mechanical impact from an accident and sends a signal to the microcontroller which is connected to a buzzer. The buzzer in response to the signal triggers an alarm and simultaneously a Short Message Service (SMS) and recorded phone call are sent to registered contacts in the device. Arduino ATMega328P microcontroller monitors and activates the system when there is impact detected by the ultrasonic sensor. Then the buzzer outputs the warning sound to alert while the Global System for Mobile communication (GSM) module sends SMS alert and places a call when there is significant impact on the vehicle. This device senses any Izod Impact causing up to 10cm penetration at the impact zone of the vehicle. The system has a response time of 0-3 seconds for activation of the buzzer and 5-10 seconds for the GSM module.
Uzoma Ifeanyi Oduah, Onyedikachi Alanwoko and Shola Folarin
We hereby present an advanced road vehicle safety device. The device tackles the problem of loss of lives, arising from late medical attention to care for accident victims after an accident as a solution that enables fast response...
