Bone fractures are a leading cause of morbidity and mortality worldwide. In Uganda, statistics for the prevalence of bone fractures are unknown, although anecdotal evidence points to a high incidence, mostly arising out of traffic accidents and falls. The situation becomes worse year on year due to a rising life expectancy, and thus an increasing number of the aging population who are more prone to fractures. To reduce the debilitating e↵ects of these fractures and improve quality of life, it is important that the fractures are accurately diagnosed early on. X-ray imaging is the most common imaging modality for fracture diagnosis in Uganda, but its manual interpretation is usually error-prone, potentially leading to missed diagnoses. To address this challenge, this project aimed at developing an automated bone fracture detection system for the efficient diagnosis, utilizing a deep learning approach. Images of fractured bones were obtained from Roboflow and the open-source dataset. We developed a model for localization of the bone fractures, utilizing the YOLOv5 architecture. Our best model achieved a mean average precision of 85.6%. Comparison with alternative approaches such as EfficientDet, and Detectron2 reveals the superior performance of our model. Our model, when integrated into a clinical decision support system, is potentially a promising approach to improve clinical outcomes based on accurate and efficient bone fracture detection from x-ray data.

Recently, localization in Global Positioning System (GPS) denied environments has witnessed an increase in interest, due to the potential wide range of using in different applications, such as Internet of Things (IoT) providing a solution for the absence of GPS signals inside buildings and tunnels. Different techniques have been used for performing this type of localization, such as sensors and wireless technologies. In this paper, localization and object tracking system is proposed based on WiFi transmission technique. It is done by deploying a WiFi access point with an Unmanned Aerial Vehicle to fly autonomously in such places to read the data of the tracked objects. This is to measure the received signal strength and distance between the WiFi access point and the object to allocate and track it efficiently. The test results show that the proposed system is working in an efficient way with low cost.

Uganda is a low income developing country with a GDP growing at a rate of 3.4%, against an inflation of 14% (World Bank, 20013). The construction industry contributes over 12% of Uganda’s GDP and has witnessed steady growth for the last 20 years and despite recent upsurge in inflation Uganda National Commission for UNESCO (2013). The problems the industry faces are still a downside to the rate of growth. The World Bank (1984) summarizes some of these problems in the developing countries as inexperienced and excessively rigid contract supervision, inadequate training staff, construction business proprietors who tend to outgrow their capacity to manage construction risk and inadequate procurement and contracting procedures leading to delayed payments without adequate compensation for contractors. Contracting procedures include the choice of Project Delivery Methods (PDM) and financial contract types (FCT) to be used for a project. A Project delivery method is a system for organizing and financing design, construction, operations and maintenance activities and facilitates the delivery of a good or service, Miller (2000). Project Delivery Methods have evolved over the years. The master builder was hired by an owner to design, engineer and construct an entire facility. This system was common until the 20th century (Sanvido, 1998)

Unmanned Aerial Vehicles (UAVs), commonly referred to as drones, have recently seen an increased level of interest as their potential use in same-day home delivery has been promoted and advocated by large retailers and courier delivery companies. The challenge is the deployment of a large fleet of unmanned aerial vehicles (UAVs) or drones currently to perform deliveries of multiple packages within a given area of delivery, with each drone carrying a single package. Moreover, for deliveries within a small neighborhood, a single delivery drone makes multiple flights to and fro the distribution center e.g., a truck or van, warehouse, depot et al, each time carrying and delivering a single package. Such methods of deployment incur high costs of last-mile deliveries, longer time of deliveries and lead to an increase in the traffic density of a given airspace. These tradeoffs must be minimized to promote the feasibility of package/parcel deliveries by drones. We introduce a novel way to exploit drones in same-day home delivery designs: multiple package deliveries. We explore a technique that can be used to support multiple package/parcel deliveries by a single multirotor (rotary wing) drone on a single flight from the distribution center/depot/warehouse to different delivery locations within a given area or radius. We introduce the last-mile delivery problem in Section 1 and present the background of the project, the problem statement, its objectives, scope and significance/justification. In Section 2, we give a detailed literature review exploring the work that has already been done in relation to multiple package deliveries by drones, a detailed review of the best contributions to multiple package deliveries and a synthesis of new knowledge from them. Finally, in Chapter 3, we present the project plan comprising the requirements specification (containing detailed description of the proposed design), technical specification (containing flowcharts for the control sequences and a schematic block diagram for the proposed design) and results/outcomes of the project.

Traditional screening of cervical cancer type classification majorly depends on the pathologist’s experience, which also has less accuracy. Colposcopy is a critical component of cervical cancer prevention. In conjunction with precancer screening and treatment, colposcopy has played an essential role in lowering the incidence and mortality from cervical cancer over the last 50 years. However, due to the increase in workload, vision screening causes misdiagnosis and low diagnostic efficiency. Medical image processing using the convolutional neural network (CNN) model shows its superiority for the classification of cervical cancer type in the field of deep learning. This project proposes deep learning CNN architectures to detect cervical cancer using the colposcopy images. The project was divided into two tasks: task one where we have a dataset containing images of three types of cervixes that is type 1, type 2 and type 3 and task two were we have a dataset containing images of cancerous (positive) and non-cancerous cervixes (negative). The datasets where obtained using a colposcope. In task one the dataset was trained using deep learning CNN architectures; Yolo V5 and Yolo V4. In dataset one the Yolo V4 gave a better performance with a mAP of 0.646 the compared to the Yolo V5 which gave mAP of 0.283. In task two we trained the classification model and the object detection model. Under the classification model the xception model had a better performance with a training accuracy of 97.13%, validation accuracy of 89.01% and test accuracy of 91.3% compared to the Inception V3 model which gave a training accuracy of 88.21%, validation accuracy of 78.1% and testing accuracy of 75.9%. under the object detection model, we only trained using the Yolo V4 which gave mAP of 0.879424. In conclusion, with more datasets more GPU time the accuracy level of our model can be improved.

Agriculture is the major economic activity carried out in Uganda as it employs more than 62% of her population either directly or indirectly. Basing on this fact, agriculture has been prioritised as a major sector which would help to elevate her economic status from lower income status to middle income status by the year 2020 through different programmes for example NAADS, and infrastructural development. However, despite all this effort by the government her ambitions to attain middle class income status by 2020 seems to still be a far cry as stated by minister of finance during the 2019-2020 budget speech. So, this research oughts to find out the role land markets have played in hindering the goal of attaining middle income status through assessing the impact of land markets on land investment singling out agriculture. From data obtained from questionnaires administered, interviews and data from national databases, it is shown that land markets have had an impact on agricultural investment as approximated by: - 1) Relationship between capital and land markets. 2) Assessing factors in land markets that affect land investments agriculture in particular. In conclusion, the research discovered that there is still dead capital (underutilisation of land as a atural resource) on land due to deficiencies' in the land market as observed that ny people ntend to trigger this capital through mortgaging such that they can invest more in commercial griculture but the bottle necks in the land market system hinder them. The government should therefore priorities streamlining the land market before it can call on vestors to invest in commercial agriculture and industrialisation.

The optimal placement of multiple unmanned aerial vehicles (UAVs) equipped with directional antennas acting as wireless base stations that provide coverage to several ground users is analyzed. In this research, the optimal placement of unmanned aerial base stations that provide coverage for the ground terminals (users) is analyzed and uses the case study of an MTN malfunctioning terrestrial base station where UAVs are deployed to give coverage to the ground users as the base station problem is being rectified by the maintenance MTN engineers First, the downlink coverage probability as a function of altitude and antenna gain is derived based on the probabilistic line of sight or non-line of sight links (Los/NLoS) links. Next, using circle packing theory, the 3-D locations of the UAVs are determined in the way that the total coverage area is maximized while maximizing the coverage lifetime of the UAVs. Given a desired geographical area that needs to be covered by multiple UAVs, an efficient deployment approach is proposed based on the circle packing theory that leads to maximum coverage while each UAV uses a minimum transit power. The results show that the optimal altitude and locations of the UAVs can be determined based on the number of available UAVs, the antenna gain, and beamwidth. Results also show that to mitigate interference, the altitude of the UAVs must be properly adjusted based on the beamwidth of the directional antenna as well as the coverage requirements. Furthermore, the minimum number of UAVs required to guarantee a target coverage probability for a given area is determined.

This report contains information on the design and construction of a pasta making machine from locally available materials. This pasta making machine is a rotary type consisting of four major components each made of stainless steel and it is manually operated. It works on the principle of a combination of crank handle rotation and extrusion of infed pasta as the output. Most of Ugandans do not have pasta maker machines, be it homes or restuarants and hotels and always have to incur a high cost of purchase of pasta and its other products from markets which are imported from foreign countries and in addition most of the existing machines are time and high-power consuming. This project report contains basic engineering terms with assumptions made and standard formula. In addition, there are basic principles of mechanical engineering which were used in design of major components of the pasta maker. It contains five chapters with each chapter well organized. References, a budget and a project time schedule were some of the information contained in project report.