Browsing by Author "Mabirizi, Vicent"

Now showing 1 - 10 of 10
  • Thumbnail Image
    ItemOpen Access
    A Mobile Based Technology to Improve Male Involvement in Antenatal Care.
    (Kabale University, 2024) Muhoza, B. Gloria; Ssemaluulu, Paul Mukasa; Mabirizi, Vicent
    The World Health Organization Technical Working Group on maternal health unit recommended a minimum level of care to be four visits throughout the pregnancy for pregnant mothers [1]. The first visit which is expected to screen and treat anaemia, and syphilis, screen for risk factors and medical conditions that can be best dealt with in early pregnancy and initiate prophylaxis if required (e.g., for anaemia and malaria) is recommended to be made before the end of the fourth month of pregnancy. The second, third and fourth visits are scheduled at 24–28, 32 and 36 weeks, respectively. Male involvement in Antenatal health care has been described as a process of social and behavioural change that is needed for men to play more responsible roles in maternal health care to ensure women's and children’s wellbeing. A study by Okoth [1] reported that, in Uganda, male involvement in antenatal care stands at only 6% and this has been attributed to social, economic and cultural related factors. The situation worsens with the lack of an effective coordinated platform for males sharing their experience in taking part in ANC and this has affected the process of antenatal care service delivery. Objective. To assess the role of mobile technology in improving male involvement in antenatal care by developing a mobile-based technology which sends SMS reminders to male partners encouraging them to escort their pregnant wives for antenatal care services. Research questions. What are the challenges towards the limited antenatal care-seeking behaviours among pregnant mothers? What are the causes of limited male involvement in antenatal care? What roles do ICTs play in enhancing Antenatal Care seeking behaviours among pregnant mothers and in increasing the male involvement in Antenatal Care? Method. We purposively selected pregnant mothers whose phones, had been receiving antenatal care services from Kabale General Hospital and reported staying with their male partners. The recruited participants were interviewed together with their male partners. STATA 13 software was used to define participants’ demographics while qualitative data were analysed using content analysis to come up with classes describing participants’ perceptions. Results. Participants reported that reminding them of their next antenatal visit via SMS reminder plays a significant role towards their antenatal care-seeking behaviour. Conclusion. Mobile health could be a potential approach to improving male involvement in antenatal care through sending timely SMS reminders to both the expectant mother and her male partner reminding them of their next antenatal visit.
  • Thumbnail Image
    ItemOpen Access
    An algorithm to Detect Overlapping Red Blood Cells for Sickle Cell Disease Diagnosis.
    (Kabale University, 2024) Mabirizi, Vicent; Kawuma, Simon; Safari, Yonasi
    In Africa, Uganda is among the countries with a high number of babies (20,000 babies)born with sickle cell, contributing between 6.8% of the children born with sickle cell every year worldwide and approximately 4.5% of the children born with hemoglobinopathies worldwide. It is estimated that by 2050, sickle cell cases will increase by 30% if no intervention is put in place. To facilitate early detection of sickle cell anemia, medical experts employ machine learning algorithms to detect sickle cell abnormality. Previous research revealed that algorithms for recognizing the shape of a sickle cell from blood smear by fractional dimension, cannot detect sickle cells if applied to blood samples containing overlapping red blood cells. In this research, the authors developed an algorithm to detect overlapping red blood cells for sickle cell disease diagnosis. The algorithm uses canny edge and double threshold machine learning techniques and takes overlapping red blood cell images as inputs to detect if these cells are sickle cell anemic. These images have a scale magnification of (200×, 400×, 650×) pixel taken using a microscope. The algorithm was tested on a total of 1000 digital images and the overall accuracy, sensitivity, and specificity were 98.18%, 98.29%, and 97.98% respectively.
  • Thumbnail Image
    ItemOpen Access
    An Empirical Study of Bugs in Eclipse Stable Internal Interfaces.
    (Kabale University, 2024) Simon, Kawuma; Nabaasa, Evarist; Bamutura, David Sabiiti; Mabirizi, Vicent
    The Eclipse framework is a popular and widely used framework that has been evolving for over a decade. The framework provides both stable interfaces (APIs) and unstable interfaces (non-APIs). Despite being discouraged by Eclipse, application developers often use non-APIs which cause their systems to fail when ported to new framework releases. Previous studies showed that applications using relatively old non-APIs are more likely to be compatible with new releases compared to the ones that used newly introduced non-APIs. Furthermore, from our previous study about the stability of Eclipse internal interfaces, we discovered that there exist 327K stable non-API methods as the Eclipse framework evolves. In the same study, we recommended that 327K stable non-API methods can be used by Eclipse interface providers as possible candidates for promotion to stable interfaces. However, since non-APIs are unsupported and considered to be immature i.e., can contain bugs, to this end, there exists a need to first investigate the stable non-APIs for possible bugs before they can be promoted to APIs. In this study, we empirically investigated the stable non-API for possible bugs using the Sonarqube software quality tool. We discovered that over 79.8% of classes containing old stable non-API methods have zero bugs. Results from this study can be used by both interface providers and users as a starting point to analyze which interfaces are well tested and also estimate how much work could be involved when performing bug fixing for a given eclipse release.
  • Thumbnail Image
    ItemOpen Access
    Assessing the Effectiveness of Tools Used for Lecturer and Course Evaluation in Institutions of Higher Learning: A Case Study from Ugandan Universities.
    (Kabale University, 2024) Mabirizi, Vicent; Karungi, Monica; Murangira, Jones; Muhoza, B. Gloria; Mutebi, Michael; Mbago, Ronald; Kohabohebwa, John Ivan; Birungi, Ruth
    Blended learning, a pedagogical method integrating face-to-face and online instructions methodologies, has been identified as a strategic educational approach since its inception in the late 1990s. Moreover, its adoption especially in developing countries such as Uganda was widely recognized during the COVID-19 pandemic’s acceleration of digital learning adoption. However, this adoption has paused many challenges in evaluating learning content, teaching methodologies, and their impact on student progress. This study therefore, explores the critical role of quality assurance in higher education, focusing on the assessment of lecturer performance and course content. Apparently, paper-based mode of evaluation is the commonly used method in Ugandan universities, posing issues of privacy, delayed analytics, and ever-increasing operational costs. To address these challenges, this research proposes the development of an automated assessment system, informed by a benchmarking study across four universities. By adopting insights from existing evaluation practices, the proposed system aims to enhance the efficiency, accuracy, and students’ privacy during lecturer and course assessment. The implementation of this automated system at Kabale University promises to streamline evaluation process, ultimately enhancing teaching quality and academic outcomes.
  • Thumbnail Image
    ItemOpen Access
    Comparison of Deep Learning Techniques in Detection of Sickle Cell Disease.
    (Kabale University, 2023) Mabirizi, Vicent; Kawuma, Simon; Kyarisiima, Addah; Bamutura, David; Atwiine, Barnabas; Nanjebe, Deborah; Oyesigye, Adolf Mukama
    Recently, the transfer learning technique has proved to be powerful in enhancing the development of deep learning methods for sickle cell disease (SCD) detection as a complement to the clinical method where a hemoglobin electrophoresis machine is used. This is evidenced by some models and algorithms with ≥90% prediction accuracy. From the literature, most of the proposed methods are trained and tested on pre-trained deep learning models like VGG16, VGG19, ResNet, Inception_V3, and ReNet. However, training and testing of these methods are limited to one model and separate datasets which may lead to biased results due to implementation in a variation of these models which affects the results produced. To this end, there exists a need to evaluate the SCD models using the same dataset. Thus, in this research study, we carried out a comparative investigation and evaluated predominate pre-trained models used to detect SCD using the same dataset to ascertain which one has the best accuracy. We used a secondary dataset obtained from an online dataset. In our study, we have discovered that Inception V3 yielded the highest accuracy of 97.3% followed by VGG19 at 97.0%, VGG16 at 91%, ResNet50 at 82% and ReNet at 67%, and the CNN-scratch model achieved 81% accuracy. Results from our study will aid researchers and industry practitioners in making decisions on the best deep-learning model to use while detecting SCD.
  • Thumbnail Image
    ItemOpen Access
    Deep Learning Techniques in DICOM Files Classification: A Systematic Review
    (BON VIEW PUBLISHING PTE.LTD., 2025) Mabirizi, Vicent; Kawuma, Simon; Natumanya, Deborah; Wasswa, William
    The digital imaging and communications in medicine (DICOM) format is a widely adopted standard for storing medical imaging data, integrating both image and metadata critical for clinical diagnostics. However, its complexity poses challenges for deep learning applications, particularly in extracting and processing this dual-layered data. This review analyzes 23 peer-reviewed studies published between 2014 and 2024, sourced from PubMed, Google Scholar, PLOS, Science Direct, and IEEE databases. Guided by Arksey and O’Malley’s scoping methodology, the review reveals that existing deep learning techniques typically rely on converting DICOM images into simpler formats like JPEG, TIF, or PNG for classification, a process that often results in metadata loss and reduced classification accuracy. Frameworks such as MONAI, NVIDIA Clare, SimpleITK, and OpenCV facilitate direct DICOM processing but face limitations, including overfitting, challenges with data heterogeneity, and inefficiencies in handling large datasets. This review emphasizes the urgent need for developing a robust convolutional neural network architecture capable of directly processing DICOM data to preserve metadata integrity and enhance predictive performance, paving way for more reliable and scalable medical imaging solutions.
  • Thumbnail Image
    ItemOpen Access
    Diagnosis and Classification of Tuberculosis Chest X-ray Images of Children Less Than 15 years at Mbarara Regional Referral Hospital Using Deep Learning.
    (Kabale University, 2024) Kawuma, Simon; Kumbakumba, Elias; Mabirizi, Vicent; Nanjebe, Deborah; Mworozi, Kenneth; Mukama, Adolf Oyesigye; Kyasimire, Lydia
    Tuberculosis (TB) is an underestimated cause of death in children, with only 45% of cases correctly diagnosed and reported. It is estimated that 1.12 million TB cases occurred among newborns, children, and adolescents aged less or equal 14 years. In Uganda, TB prevalence is 8.5% in children and 16.7% in adolescents. Treatment and diagnosing TB is challenging and its high mortality rate is due to many lacks in the diagnosis of this illness especially among children. As a strategy to curb TB mortality rate in children, there exists a need to improve and expedite the screening for TB among children. Chest X-ray (CXR) is commonly used in TB burdened countries like Uganda to diagnose TB patients but interpretation of the patient’s radiograph needs skilled radiologists who are few. To this end, this research aims to close the TB mortality gap in children by applying AI, primarily deep learning techniques, to detect TB in children. The study created five models, one from scratch and four pre-trained Transfer Learning (TL) and were trained and verified using digital CXR radiograph images of children who visit the TB clinic at Mbarara Regional Referral Hospital. The model classifies clinical images of patients into normal or Tuberculosis. TL models; VGG16, VGG19, Inception V3, and ResNet50 outperformed scratch model with validation accuracy of 79.91%, 69.21%, 53.0%, 51.09% and 50.01% respectively. We hope that once the deep learning models are implemented and adopted by the radiologist, it will reduce the time spent by radiologist while analysing CXR images.
  • Thumbnail Image
    ItemOpen Access
    Green gram yield prediction using linear regression
    (African Journal of Rural Development, 2025) Tumusiime, Robert; Mabirizi, Vicent; Mirembe, D.P; Arinanye, T.R
    Predicting crop yields before harvest is key in enabling farmers make critical decisions as far as postharvest management is concerned. Besides, yield prediction plays a critical role in agriculture enterprise selection hence promoting food and nutrition security in a community. It is worth noting that various factors including ecological zones characteristics and farm management practices can vary significantly from season to season and farm to farmer, hence affecting crop yields. Given the importance of crop yield prediction in agriculture enterprise development and investments, a number of approaches have been adopted by farmers and breeders alike. These approaches range from controlled ideal condition analysis by breeders to the use of advanced plant physiological feature analysis using satellite image processing techniques. While a number of popular crops like rice and maize have a number of models proposed, limited yield prediction studies have been done on neglected crops like green gram. Therefore, this paper discusses the proposed green gram crop yield prediction model based on a stepwise linear regression technique using ecological zone characteristics, farm management practices and historic crop yield as the key variables. The study used a dataset of 107 records (gardens) and 9 features obtained from National Semi-Arid Research Institute (NaSARRI), Serere, Uganda. The predictor variables used were; soil type, soil PH, soil fertility, rainfall distribution, weeding practice, pest and disease management, fertilizer application, plant spacing, and cropping system. The model was evaluated for precision and evaluation result revealed that, with a mean absolute percentage error (MAPE) of 16.8%, the proposed model had a precision of 96.4% was deemed accurate in predicting green gram yield.
  • Thumbnail Image
    ItemOpen Access
    Masked and Unmasked Face Recognition Model Using Deep Learning Techniques. A case of Black Race.
    (Kabale University, 2023) Mabirizi, Vicent; Ampaire, Ray Brooks; Muhoza, Gloria B.
    Currently, many institutions of higher learning in Uganda are faced with major security threats ranging from burglary to cyber threats. Consequently, the institutions have recruited and deployed several trained personnel to offer the desired security. As human beings, these personnel can make errors either by commission or omission. To overcome the limitation of trained security personnel, a number of face recognition models that detect masked and unmasked faces automatically for allowing access to sensitive premises have been developed. However, the state -of -the art of these models are not generalizable across populations and probably will not work in the Ugandan context because they have not been implemented with capabilities to eliminate racial discrimination in face recognition. This study therefore developed a deep learning model for masked and unmasked face recognition based on local context. The model was trained and tested on 1000 images taken from students of Kabale University using Nikon d850 camera. Machine learning techniques such as Principal Component Analysis, Geometric Feature Based Methods and double threshold techniques were used in the development phase while results were classified using CNN pre-trained models. From results obtained, VGG19 achieved the higher accuracy of 91.2% followed by Inception V 3 at 90.3% and VGG16 with 89.69% whereas the developed model achieved 90.32%.
  • Thumbnail Image
    ItemOpen Access
    A Mobile Based Technology to Improve Male Involvement in Antenatal Care.
    (Kabale University, 2022) Gloria, Muhoza B.; Mukasa, Ssemaluulu Paul; Mabirizi, Vicent
    The World Health Organization Technical Working Group on maternal health unit recommended a minimum level of care to be four visits throughout the pregnancy for pregnant mothers [1]. The first visit which is expected to screen and treat anaemia, syphilis, screen for risk factors and medical conditions that can be best dealt with in early pregnancy and initiate prophylaxis if required (e.g., for anaemia and malaria) is recommended to be made before the end of the fourth month of pregnancy. The second, third and fourth visits are scheduled at 24–28, 32 and 36 weeks, respectively. Male involvement in Antenatal health care has been described as a process of social and behavioral change that is needed for men to play more responsible roles in maternal health care with the aim of ensuring women and children’s wellbeing. A study by Okoth [1] reported that, in Uganda male involvement in antenatal care stands at only 6% and this has been attributed to social, economic and cultural related factors. The situation worsens with the lack of effective coordinated platform for males sharing their experience in taking part in ANC and this has affected the process of antenatal care service delivery. Objective. To assess the role of mobile technology to improving male involvement in antenatal care by developing a mobile based technology which sends SMS reminders to male partners encouraging them to escort their pregnant wives for antenatal care services. Research questions. What are the challenges towards the limited antenatal care seeking behaviours among pregnant mothers? What are the causes of limited male involvement in antenatal care? What roles do ICTs play in enhancing Antenatal Care seeking behaviours among pregnant mothers and in increasing the male involvement in Antenatal Care? Method. We purposively selected pregnant mothers who phones, had been receiving antenatal care services from Kabale general hospital and reported staying with her male partner. The recruited participants were interviewed together with their male partners. STATA 13 software was used to define participants’ demographic while qualitative data were analysed using content analysis to come up with classes describing participants’ perceptions. Results. Participants reported that reminding them of their next antenatal visit via SMS reminder plays a significant role towards their antenatal care seeking behaviour. Conclusion. Mobile health could be a potential approach to improving male involvement in antenatal care through sending timely SMS reminders to both the expectant mother and her male partner remaining them of their next antenatal visit. Keywords: Antenatal Care, Male Involvement, Mobile Health Care, Digital Health.