Mak IR Repository (Makerere University)
Not a member yet
9127 research outputs found
Sort by
Congestion Control for QUIC protocol using RNN And LSTM as deep learning models (COCO-QUIC)
No full textA Dissertation Submitted to the Directorate of Research and Graduate Training for the Award of the Degree of Master of Science in Data Communication and Software Engineering of Makerere University.QUIC (Quick UDP Internet Connections) is a new transport protocol that was designed on the motivation of improved speed, better performance and enhanced security. This protocol has been adopted in several use IoT cases and applications. Several research efforts have demonstrated performance improvements by switching to QUIC however, new challenges continue to come up. As Internet networks grow due to massive demand attributed to the Internet of Things (IoT), network congestion continues to cause huge challenges along the transport Layer. Tremendous efforts, including deep learning methods, have been used to improve network congestion over time. Although deep learning methods have been used to address network congestion challenges in many works, less research has been directed towards the QUIC protocol, therefore, this study proposes Congestion Control for QUIC using Recurrent Neural Network (RNN) and Long-Short Term Memory (LSTM) as Deep Learning
models (COCO-QUIC). The datasets for this study were collected usingWireshark packet sniffer from Makerere University Network Operations Center (NOC) and contains 1,549,088 samples of records with parameters including e-mail, video streams, audio streams, social media, browsing, and anomaly-related traffic. Our main objective is to develop a congestion control model and two supervised deep learning models have been stacked together and used in this study which are the Simple Recurrent Neural Networks (Simple RNN) and Long-Short Term Memory (LSTM), formulating COCO QUIC model. The Simple RNN model
has been used as the initial model with the dataset for training, testing, and validation and achieved an overall accuracy of 92.57% on the collected data samples, the LSTM model attained an overall accuracy of 92.90% on the same collected data samples and COCO QUIC attained an overall accuracy of 93.33%. The two stacked models of LSTM and RNN were compared with RNN,LSTM and RF as existing models for performance evaluation and were found to perform much better with less Latency, Delay, Packet loss and improved bandwidth. The new COCO QUIC model performed better than Simple RNN, LSTM, and RF, indicating that the designed model show improvement in performance on congestion compared to existing models. This research
study’s contribution is the performance of COCO QUIC model on congestion control measures of different learning rates applied to the Simple RNN, LSTM and R.F models on parameters such as throughput, delay, packet loss, latency and it was also observed that COCO QUIC performed much better than Simple RNN, LSTM and RF
Role of society organizations in the promotion of the rights of male victims of inter-partner violence in Uganda: the case of relief teenage and marital counseling Uganda
No full textA dissertation submitted to the Directorate of Research and Graduate Training in partial fulfilment of the requirement for the award of the Degree of Master of Arts in Human Rights of Makerere UniversityThe purpose of this study was to find out the role of Civil Society Organizations in Uganda in the promotion of the rights of male victims of Inter-Partner Violence (IPV) in Wakiso district using the case of Relief Teenage and Marital Counseling Foundation Uganda (RETMAC-F). The study sought to identify the factors leading to the prevalence of IPV, the diffrent forms of IPV faced, roles played by RETMAC in the promotion of rights of IPV victims and the challenges faced by RETMAC in trying to promote the rights of IPV victims. The study was mainly conducted using a qualitative method of interviews. Data was collected from male victims, RETMAC staff, Local Council 1 members and police family desk officials. The study established the forms of IPV as; psychological, sexual and physical and these include; being pushed and locked out of house, throwing dangerous items such as glasses and clay plates, kicking, cutting using machete and struggling; involvement in extra marital sex and denial of conjugal rights; verbal abuse, stalking, humiliation and defamation in front of the public, control and intimidation, treating the men like children and isolating the husbands from the family. In addition, the factors leading to the prevalence of IPV according to male victims were; taking alcohol, smoking, having multiple wives, poverty, family background, low level of education of the female partner, family neglect, unemployment, drug abuse and sexually transmitted diseases. The role played by RETMAC include; paying medical bills to the victims who are physically injured, re-uniting male victims, information sharing on IPV against men, conducting outreach campaigns and radio talk shows, increasing men’s capacity to report IPV, creating a violence free home environment and establishing a rehabilitation center for battered men and women. The challenges are; non-disclosure of IPV cases among men, attacks from competing human rights activists or defenders, limited financial resources, and limited reliable data on the prevalence of IPV and sustainability issues. The study concluded that IPV against men still exists in Wakiso district though less than IPV against women. The study recommended that a similar study should be conducted in the whole of central region to understand the overall prevalence and the factors leading to IPV against men such that a framework for complete elimination of IPV against men can be designed. Also, the study conducted should incorporate statistics to help planners to mobilize resources enough to implement a successful intervention
Evaluation of the performance of the newborn sickle cell screening program and associated facilitators and challenges in Uganda
No full textIntroduction
In Africa, scaling up of national screening programs has been shown to be difficult. This could be attributed to the various challenges facing these programs in Africa. Therefore, is important to monitor progress of screening programs in order to identify areas for improvement.
In Uganda, currently there has not been any study to establish the performance in implementation of the new born sickle cell screening program. Therefore, there is a gap in evidence as regards the performance in implementation of the program. This study aims to assess the performance of implementation of the program based on the number of children screened within the age of 3months and explore the challenges and facilitators associated with implementation of the program in Uganda.
Objective: This study aimed to evaluate the performance of the newborn sickle cell screening program and explore the associated facilitators and challenges in Uganda.
Methods
A mixed methods study based on secondary data from the HMIS 105 and key informant interviews of program staff based in different health facilities to determine the extent of performance of the sickle cell program and associated challenges and facilitators. Performance was determined by the standards set by the Consortium on Newborn Screening in Africa for sickle cell disease. Secondary data for the period July 2021 to Jun 2023 was obtained from the national HMIS 105 database and analyzed using STATA. A pretested modified questionnaire was used to collect data on the challenges and facilitators. The qualitative data was collected from hospitals including Mengo Hospital, Lira Regional Referral Hospital, Mulago National Referral Hospital and Entebbe Hospital. This data was then analyzed using thematic analysis using NVivo.
Results: The findings indicate that for the periods Jul 2021 to Jun 2022 and Jul 2022 to Jun 2023, only 1440 and 1900 children were screened within three months’ age respectively. These were below the CONSA target of 10,000 children per annum. There was variation in the proportions of children screened within 3 months’ age across regions with the South western region (9%) having the highest percentage followed by central 2 region with 7.4%, central 1 region with 3.8% then Kampala with 3.5%. The lowest proportion observed in West Nile and East Central each at 0.4% each. Statistical analysis using chi square test revealed a p value of <0.005.
Of all the children screened within the age of 3 months across the periods Jul 2021 to Jun 2022 and Jul 2022 to Jun 2023, 71.9% were male compared to 28.1% who were female. Statistical analysis using chi square test revealed a p value of 0.0026. This indicates a statistically significant difference between the male and female children screened within 3 months’ age.
The major challenges of the program were centered on input issues such as shortage of supplies which in turn lead to process issues such as inadequate motivation and mobilization of health workers to participate in the program and poorly filled and incomplete forms. The major facilitators included government support such as supply of hydroxyurea and effective transportation of samples and dissemination of results coordinated by CPHL.
Discussion and recommendations: Screening of children within the age of three months is still a challenge due to input issues evidenced by the lack of resources and process issues such as incomplete forms and poor sample handling. These can be addressed by adjusting the age at which screening is done to from within 3months’ age to within age of 1 year and integration of the screening program into existing antenatal and post-natal care health services such as immunization could ensure that training, mobilization and other process activities run seamlessly.
Conclusion: The numbers of children screened within three months’ age are below the CONSA target. The proportions of children screened within three months’ age vary with regard to region which indicates inconsistency in the implementation of the program in the various parts of the country. This is because of input issues and the fact that the neonatal screening was warranted in districts with the highest burden.
According to this study, there are also more male children within the age of three months being screened compared to the female children within the same age. There is a need for more research to investigate the reason this observation given that this study did not involve the beneficiaries of this program.Health Professions Education and Training for Strengthening the Health System and Services in Uganda (HEPI-SHSSU) Grant No. 1R25TW01121
Risk of respiratory distress syndrome in preterm neonates born to mothers who receivec antenatal dexamethasone between 2 weeks 0 days and 33 weeks days at kawempe national referral hospital
No full textA dissertation submitted to the directorate of research and graduate training in partial fulfillment of the requirements of the award of master of medicine in obstetrics & gynecology of Makerere UniversityIntroduction: Respiratory distress syndrome (RDS) is a serious complication of preterm birth and is the main cause of perinatal mortality. Corticosteroids have reported to have a positive effect on the mother and fetus with minimal secondary ill effects because they bind to 75 to 80% of alveolar glucocorticoid receptors, sufficient to stimulate lung function. However, mothers should be given full course of glucocorticoids for them to be protected from the risk of RDS. This study aimed at determining whether a partial course of dexamethasone is protective of RDS and to determine obstetric factors associated with RDS among preterm neonates. Methods: This was a case-control study conducted at Kawempe National Referral Hospital. Data was collected from preterm neonates born at (26W0D and 33W6D) admitted in the NICU with RDS as cases and those without RDS as controls and thereafter establish the exposure status (whether their mothers received partial or full dose of dexamethasone). The sample size of the study was 375 which included 125 case patients and 250 matched controls. Data was analyzed using STATA version 15. Association between variables was established using logistic regression analysis, the measure of association wask odds ratio at 95% level of confidence interval. Results: The average age of the respondents was 26.13 (±4.95) and majority were married 313 (83.47%). Only 20 (5.33%) mothers (for both cases and controls) were given a full course of dexamethasone as majority 194 (51.73%) were given only one dose. After adjusting for potential cofounders, RDS was found to be significantly associated with partial courses of dexamethasone (AOR=2.31, P-Value 0.031). The risk of RDS was also found to be high in preterm neonates who were twin deliveries (AOR=5.3, P-valu
Machine learning and orthology-based design of synergistic drug combinations against aspergillus fumigatus
No full textA dissertation submitted to the Directorate of Research and Graduate Training in partial fulfillment of the award of the degree of Master of Science in Bioinformatics of Makerere UniversityBackground: Aspergillus fumigatus is a common fungal species that can cause invasive aspergillosis, particularly in immunocompromised individuals. In Uganda, around 9% of the population suffers from fungal diseases. Drug resistance and limited access to antifungals in low-income countries present major challenges. While drug development and vaccines offer potential solutions, they are often costly and time-consuming. To address this, the study utilized chemogenomic data from Cryptococcus neoformans and applied machine learning to identify synergistic drug combinations against A. fumigatus, using C. neoformans as a model and mapping orthologous genes to extrapolate drug interaction profiles to A. fumigatus. Methods: Large-scale datasets for Cryptococcus neoformans including chemogenomic data, antifungal drug structures, and Minimum Inhibitory Concentrations (MICs) from checkerboard assays were collected and used to train machine-learning models after performing a series of data preprocessing steps. Machine learning algorithms like random forest and gradient boosting classifiers and ensemble methods were used to develop predictive models that identified patterns between C. neoformans and antifungal drugs, as well as orthologous genes from Aspergillus fumigatus identified via OrthoFinder. Results: The Gradient Boosting Classifier demonstrated the highest accuracy at 82%, with an AUC of 70% and a Matthews Correlation Coefficient of 0.71. The study also identified 276 synergistic drug combinations, including those involving 2-aminobenzothiazole, which showed potential when combined with FK506, 5-fluorocytosine, and 4-hydroxytamoxifene. Conclusion: The MOSCAF model successfully integrated chemogenomics, orthology mapping, drug structural information, and machine learning to identify synergistic drug combinations for Cryptococcus neoformans and Aspergillus fumigatus. The model achieved an AUC of 0.81, effectively predicting combinations involving 2-aminobenzothiazole, demonstrating its potential in accelerating antifungal drug combination discovery and addressing drug-resistant fungal infections through bioinformatics-driven approaches.Infectious Disease Institute. Genomics project(333
Effect of Treatment delay of cervical cancer on quality of life of patients at the Uganda Cancer Institute
No full textA dissertation submitted to the School of Medicine in partial fulfilment of the requirements for the award of a Master’s Degree of Medicine in Obstetrics and Gynaecology of Makerere University.Introduction: The burden of cervical cancer represents a major growing public health concern globally with an estimated age-standardized incidence of 56.2 cases per 100000 women years and a mortality rate of 7.2 deaths per 100000 women years. Cervical cancer is a preventable disease with early diagnosis and treatment having shown to improve survival rate, and quality of life and reduces mortality and morbidity. However, there is little information regarding the time interval between diagnosis and treatment initiation of cervical cancer patients in Uganda and how treatment delays can influence patient quality of life. This study aimed to evaluate the effect of delays in cervical cancer treatment initiation on patient quality of life among Ugandan women.
Methodology: This was a cross-sectional study conducted among 456 women initiated on any form of cervical cancer treatment at the Uganda cancer institute (UCI). Data on the quality of life was collected using the Functional Assessment of Cancer Therapy with Cervical Cancer Subscale (FACT- Cx) questionnaire and analyzed using STATA version 16. To examine the effect of treatment delay on the quality of life of study participants, a chi-square test was done. Adjustment for potential confounding variables was done using logistic regression analysis. A P-Value <0.05 was considered statistically significant. Results. The average time from cervical cancer diagnosis to treatment initiation was 98.73 days (IQR 40-102), with only 53.1% starting within the recommended 60 days. Timely treatment initiation was significantly associated with religion (P=0.026), employment status (P=0.006), distance to health facilities (P=0.036), and cancer differentiation state (P=0.006). Participants' overall quality of life (QoL) score was 57.14 (SD 0.6), indicating moderate well-being. Only 9.4% reported good QoL, while 30.5% had poor QoL. No significant relationship was found between treatment delays and QoL (P=0.731). Conclusion: Cervical cancer patients in Uganda demonstrated substantial delays to initiate cervical cancer treatment after diagnosis, although this has no significant effect on their quality of life. There is need to develop clear and efficient treatment pathways to minimize delays in treatment initiation but also focus on other factors, such as psychological support, symptom management, and social support to enhance patient quality of life
The impact of risk management on credit processes of microfinance institutions in Uganda
No full textResearch submitted to the College of Business and Management Sciences in partial fulfillment of the requirements for the award of the Degree of Master of Arts in Financial Services of Makerere University.This study examines the impact of risk management in the credit processes of financial institutions in Uganda. In the world of volatile cash movements and increasing global lending and borrowing of funds, financial institutions are continually affected by borrower’s late and nonpayment of loan obligations. This result from the institution's inability to collect anticipated interest earnings as well as loss of principal resulting from loan defaults. Financial institutions carry out credit risk management as a measure of administering credit to borrowers. With well-developed credit processes and procedures, credit appraisal, training of staff and setting credit standards and terms to offset the possibility for loss and improve on financial performance.
Financial Institutions thus develop strategies to either eliminate or reduce this credit risk. In the management of this risk, financial institutions are concerned about their financial performance
The previous scholars have addressed the key issues in relation to Credit Risk and portfolio management, profitability, performance of loans and the strategies and controls to manage incidences of default and made appropriate recommendations.
While there have been much research on Credit, few researchers have taken into consideration the gaps in credit processes, failure points, breaches and its impact to financial institutions with its resultant loss.
In the study, descriptive cross-sectional survey design was used because this method is effective when gathering data of a sample population at a particular point in time. The study will employed purposive and convenient sampling techniques with primary and the secondary data, using a self-administered questionnaire (put to reliability and validity tests). Questionnaires respondents were sampled out of the financial and banking industry. However, despite the efforts made to address the poor credit risk management, financial institutions still have difficulties resulting from the credit risk management processes undertaken and changes in customer base leading to decreasing financial performance. To most financial institutions, their credit risk management goal is to arrange the institution’s financial affairs in such a way that however much borrowers fail to pay back the loan in the future
Antimicrobial susceptibility patterns of escherichia coli isolates from livestock and predisposing factors to antimicrobial resistance in Isingiro and Kamuli Districts, Uganda
No full textA dissertation submitted to the Directorate of Research and Graduate training in partial fulfilment of the requirements for the award of the Degree of Master of Veterinary Preventive Medicine of Makerere UniversityAntimicrobial resistance (AMR) is a global health challenge to human and animal population. Increase in livestock population coupled with intensive husbandry practices has greatly resulted into increased utilization of antimicrobial products in the management of livestock diseases. This was a cross-sectional study conducted between August and November 2022 in 183 and 87 randomly selected livestock keeping households in Kamuli and Isingiro Districts respectively; to determine the antimicrobial susceptibility patterns of Escherichia coli isolates from livestock. Fresh fecal samples collected from rectum of each cattle, goats, and cloacal swab from chicken were placed in Stuart Transport medium and taken to College of Veterinary medicine, Animal Resources and Biosecurity public health laboratory for processing. Escherichia coli culture and isolation was done using chromocult selective media. This was followed by antimicrobial susceptibility testing (AST) by the Kirby Bauer disc diffusion method. A total of fecal samples was collected. However, 683 fecal samples were collected from which 572 E. coli isolates were obtained. These were 191 (33.1%) from cattle, 200 (35.0%) from goats and 181 (31.6%) from chicken. Only samples from which E. coli was isolated were subjected to AST. The study revealed that E. coli isolates exhibited variation in susceptibility with highest susceptibility 97.5% to kanamycin, and resistance 79.8% to tetracycline. Correct farmers’ knowledge of AMR, human sharing water sources with livestock and open field disposal of animal wastes and use of trained veterinarians were significant predictors of AMR. Evidence of the high prevalence of AMR documented here provides a baseline for the need of further surveillance and call for urgent implementation of mitigation strategies to combat this public health challenge in line with the Uganda AMR National action plan
The impact of biotechnology on human dignity: case study of Mukono Parish, Lugazi Diocese
No full textA research paper submitted in partial fulfilment for the award of Master of Theological and Religious Studies Degree of Makerere UniversityBiotechnology has revolutionized various aspects of human life, from medicine to agriculture. However, its rapid advancement raises concerns about its impact on human dignity. While biotechnology offers immense benefits, such as improved healthcare and enhanced quality of life, it also poses significant risks, especially the one of compromising human dignity, the core of human existence. This study specifically focuses on the impact of biotechnology on human dignity and proposes a moral-theological response. The main objective is to bring to light the dark side of biotechnology towards human dignity and propose a practical approach based on the Church’s teaching about the sanctity of life. This was a case-study which used both quantitative and qualitative research methods, through in-depth interviews and questionnaires to randomly selected respondents in Mukono Parish in Lugazi Diocese. The findings of the study indicate that biotechnology is the application of technology to life. The study also shows that human dignity is an inherent worthiness of every human being regardless of the circumstances, which is inviolable and should be upheld. The findings also indicate that biotechnology’s violation of human dignity, stems in the alternation of one’s consciousness about the sanctity of life, which changes the way one chooses, thinks, and responds to different phenomena in life. The study further underscores sanctity of life, a proper understanding of the dignity of man with philosophical and theological perspectives complementing each other, and consideration of the two major dimensions of a human person; physical and spiritual being. In the same way, this study proposes formation of consciences towards the truth, proper use of freedom, acquisition of the right information, and putting good policies based on human values. The study finally recommends the augmentation of the respect for human life, the fight against manipulation, provide information about biotechnology products, moral education, and self-control
A practical UNet denoising algorithm for enhanced Malaria detection in thick blood smear images
No full textA Master’s thesis submitted to the School of Computing and Informatics Technology in partial fulfilment of the requirements for the Award of the Degree of Master of Science in Computer Science of Makerere University.This thesis addresses the challenge of enhancing malaria detection in thick blood smear images by proposing a UNet-based denoising algorithm. Noise and artifacts in these images can compromise the accuracy of malaria diagnosis. The algorithm, based on the UNet architecture, is developed to remove noise and artifacts, facilitating easier and more accurate identification of malaria parasites. Various preprocessing techniques, including median filters, mean filters, and morphological filters, are explored to mitigate prevalent noise types like speckle, Gaussian, and salt-and-pepper noise. The significance of denoising lies in its potential to minimize misdiagnoses that contribute to false positives and negatives in malaria-related cases, thereby reducing unnecessary drug administration and potential health complications. The proposed UNet denoising algorithm is trained on datasets containing both noisy and clean thick blood smear images. Evaluation against existing denoising methods demonstrates superior performance in terms of denoising quality and malaria detection accuracy. The results reveal the effectiveness of the algorithm in improving the accuracy of malaria diagnosis by effectively removing noise and artifacts from thick blood smear images. The UNet denoising algorithm achieved an average Structured Similarity Index (SSIM) of 0.92, with a minimum of 0.78 and a maximum of 0.98. When the images from the dataset with these results were fed into a malaria parasite detection model, model yielded a precision was 0.75, indicated that 75% of the identified ’Parasites’ are correct, recall of 1.00, meaning that all instances of ’Parasites’ were correctly identified and an F1-Score of 0.86 demonstrating a balance between precision and recall for the ’Parasites’ class. This thesis underscores the practicality and efficacy of the UNet-based denoising algorithm as a promising solution to improve malaria detection in thick blood smear images, offering a significant step toward more accurate and reliable diagnostics in the fight against malaria