School of Computing and Informatics

MULTI DIMENSIONAL MODEL FOR E-HEALTH IMPLEMENTATION IN PUBLIC HEALTH CARE FACILITIES: A CASE OF EASTERN UGANDA

2026-04-16T12:27:19Z

MULTI DIMENSIONAL MODEL FOR E-HEALTH IMPLEMENTATION IN PUBLIC HEALTH CARE FACILITIES: A CASE OF EASTERN UGANDA ADONG, GRACE Healthcare systems consist of mechanisms intended to provide medical care, advance public health and guarantee that everyone has access to healthcare resources. The challenges facing the public healthcare sector in developing countries are mainly associated with weak healthcare systems. Weak and poor-quality health information systems contribute to failure of information flow across integrated health systems pathways; hence there is growing need to strengthen the components of Health Information Systems. It is also agreeable that health information systems have immense benefits, yet their implementations vary greatly among different countries; hence benefits to beneficiary translational gap. It is further notable that, the adoption, diffusion, acceptance and utilization of this innovation in the healthcare context are lagging. The study addressed these concerns by examining the factors influencing eHealth in public healthcare facilities and develop a multi-dimensional model for eHealth implementation in developing countries, a case of eastern Uganda. This was achieved through a survey, encompassing both quantitative and qualitative strategies. Three healthcare facility study sites were identified through predetermined selection criteria in Eastern Uganda region. The respondents cut across the different cadres of health workers, administrator, and IT staff. The collected quantitative data was descriptively and inferentially analyzed, through different analytical techniques using STATAv17, while qualitative data was thematically analyzed. From the identified study sites, key leaders were identified to respond to interview questions for qualitative data. IT personnel were interviewed to give expert response. Further, observation protocol aided data collection which equally was analyzed qualitatively, together with documents analysis. A significance test of the model components was done from which the following results were observed social factors with p-value = 0.000, Organisational factors had a p-value = 0.012, Environmental factors with a p-value = 0.024 and Technological factors that had a p-value = 0.047. A multi dimensional model was developed from the study’s outcomes to guide and support effective eHealth implementation. The model was informed by a conceptual framework which was tested and validated using a structural equation model.

MACHINE LEARNING MODEL FOR TRAFFIC SIGN RECOGNITION

2026-04-16T12:19:36Z

MACHINE LEARNING MODEL FOR TRAFFIC SIGN RECOGNITION Prestone, Jeremiah Simiyu Autonomous (driverless) cars are increasingly becoming popular, hence calling for robust Traffic Sign Recognition (TSR) systems to ensure road safety. A report by World Health Organization Road Safety Report of 2018, shows that failure to distinguish and recognize traffic signs is among the leading causes of accidents. Existing TSR systems are adversely affected by environmental conditions, partial occlusion of traffic sign, illumination, colour deterioration because of their exposure to different rays including Ultra-Violet (UV), physical deformation, variations in pictogram designs and weather conditions among others. The study was guided by the following main objective; to develop a robust model using machine learning for recognition of traffic signs. Deductive research approaches, was used to achieve the following specific objectives: to analyses existing techniques in TSR, to design an advanced feature extraction technique for robust TSR and develop a machine learning model for recognition of traffic signs. The new Local Directional Histogram of Oriented Gradient (LD-HOG) feature extractor is the main contribution of this work. A more resilient and discriminative descriptor is produced by combining the directional resilience and noise invariance of the Local Directional Pattern (LDP) with the potent gradient magnitude representation of HOG. The German Traffic Sign Detection Benchmark (GTSRB) dataset, was used to extract features. With an average F1-score of 96.5% using an SVM classifier, LD-HOG outperformed HOG by 4.2% and LDP by 7.8%. By helping to create more accurate and dependable advanced driver-assistance systems, the study will benefit a variety of stakeholders and road users, including drivers, passengers, and legislators., and policy makers.

INTEROPERABILITY MODEL FOR ELECTRONIC MEDICAL RECORDS END TO END IMPLEMENTATION

2026-04-16T11:14:59Z

INTEROPERABILITY MODEL FOR ELECTRONIC MEDICAL RECORDS END TO END IMPLEMENTATION ELAINE, KANSIIME PAMELA The adoption and utilization of Electronic Health Records can address numerous patient care difficulties and facilitate a paperless environment in hospitals; nevertheless, their full potential remains underexploited due to significant interoperability challenges. Patients typically pursue healthcare services from diverse providers operating inside either affiliated or independent organizations. In the absence of a consistent linkage across various healthcare providers, patient medical information becomes fragmented, incomplete, and obsolete. Numerous health institutions in Kenya have adopted EHRs/EMRs; nonetheless, both the nation and the facilities that have automated their records suffer from inadequate data convergence due to interoperability challenges among the various EHR systems. The interoperability of EHRs, a complicated and challenging undertaking, is predominantly lacking in poor countries. The identified target audience encounters these problems, along with architectural and technological obstacles, in facilitating seamless communication across EHRs. The study’s main objective was achieved through four specific objectives as follows; to establish EMR services available, their utilization, level of integration in the health sector; to determine the various EMR systems state of interoperability; to determine the strategies that have been used by successful implementers to address interoperability challenges and to develop an interoperability model for EMR implementation. The researcher scrutinized the following frameworks and theories to identify their weaknesses, and thereby guide the development of the interoperability model for EMR implementation: the European Interoperability Framework (EIF), Social technical systems theory, Luhmann's Social Systems Theory and Lopez and Blobel's Interoperability Framework. To underpin the study, the researcher adopted a pragmatic philosophical standing point to guide the researcher’s worldview of the research. Further a deductive and inductive approaches were also adopted for the purpose of triangulation of data. The study population included 229 healthcare workers from which a sample of 184 respondents were obtained using stratified and simple random sampling. Key informant interviews and structured questionnaires were used to gather data. Descriptive and inferential statistics were used to analyze and evaluate the quantitative data, while thematic analysis was used for analyzing the qualitative data. The findings of the study indicate technical interoperability (p=0.000<0.05), social interoperability (p=0.006<0.05), organizational interoperability (p=0.000<0.05) and semantic interoperability (p=0.000<0.05) and legal interoperability (p=0.002<0.05) were significant predictors. From the results of the study, a model of Electronic Medical Records End to End Implementation was formulated. The study concluded that technical, social, organizational, semantic and legal, factors must be addressed to achieve interoperability of EMR end to end implementation. From the findings it is recommended that challenges such as privacy concerns, implementation costs issues must also be addressed to fully harness the benefits of EMR interoperability. The facilities to invest on security measures, capacity building, compliance to government regulation, technology as strategies to address interoperability challenges. The model developed is a basis upon which future implementation of EMR interoperability end to end implementation can be based.

INTEROPERABILITY MODEL BASED ON INTERNET OF THINGS FOR SMART HOMES

2026-04-16T11:12:33Z

INTEROPERABILITY MODEL BASED ON INTERNET OF THINGS FOR SMART HOMES Metto, Shadrack Kimutai The Internet of Things (IoT) is a group of assorted technologies working seamlessly together. The recommended approach to IoT is to support interoperability in the midst of heterogeneous devices. The challenges associated with smart homes is interoperability in areas such as message exchange, difference in protocols used, energy consumption, antenna design, how to implement adaptive techniques for dynamic situations in the face of heavily constrained resources and security. The objectives of this research are: - To determine the state of device interoperability in Smart Homes, to establish the role of interoperability in Smart Homes in the Internet of Things environments, determine the factors that affect interoperability of devices in smart homes and to develop a model for interoperability in smart homes. The research contributes to the overall concept of making smart homes a reality and enhance acceptance and deployment of heterogeneous Internet of Things devices. The scope of the study included smart home users and experts who deploy smart home devices in Kenya. Two theories adopted were the standards based theory for service providers and voluntary theory which guided the research. The population covered the users/owners and the experts/vendors or technical staff who deploy smart devices in homes. The data collection methods were interviews and observations. Purposive and snowball sampling was adopted. 18 users and 7 vendors were used. The data was analyzed to derive the mean and standard deviations using statistical package for social scientists and presented in tabular form, pie charts, bar graphs and line graphs. Critical factors found in the investigation were: - Technical, Organizational, Semantic and Legal metrics are critical in the deployment of interoperability of devices in smart homes. A model was developed and a test device implemented to prove viability of the model. The model developed meets the basic needs of the users within a heterogeneous environment however there is need for further education to users on available smart home devices solutions in view of interoperability.