Kasese District is partially a rich ecosystem with enormous ecological and natural values such as agriculture soils. Estimation of crop yields is of great essence to food security. Lake Katwe subcounty is a salty area and has been dominated by salt mining rather than agriculture. This is mostly attached to the local claim that these soils do not favour crop growing and as a result of the people lacking sufficient knowledge of the salt levels as one moves away from the lake and the crops that can be grown in such area putting into consideration its salinity levels. Remote sensing is a valuable tool in decision making and management thus can appropriately be used to study soil salinity around this area with validation of soil samples. This study aimed at assessing the impact of soil salinity on crop growing with a basis on remote sensing technology and field data. In analysis, multitemporal acquired from three different years 1987, 1995 and 2015 were used. Firstly, preprocessing was done for all images to convert digital values to reflectance values while carrying out the necessary corrections such as the atmospheric correction. Thereafter several salinity indices such Normalized Difference Salt Index, Salinity Index in comparison with the vegetation Index known as the Normalized Difference Index were computed and then analyzed accordingly. The electricity conductivity of the different measurements of the field data collected in January 2020 and the necessary laboratory measurements undertaken was used as ground truth data. Basing on the results obtained from the comparison of NDVI, NDSI, SI 1 and field measurements, it was observed that as time elapses with other factors like the sunny weather and farming methods used around the area in Lake Katwe subcounty, the spatial extent of the areas being affected by soil salinity are expanding. Conclusively, this approach is considered a very effective method and technology for assessing the impact of soil salinity on crop growing and thus can be applied to other area with similar or slightly varying characteristics.

The study was conducted in Kiboga district and the main objective was to assess rainfall variability and its impacts on crop farmers. The study analyzed historical rainfall trends from 1988 to 2018 as well as how crop farmers in the study area adapt to the impacts of rainfall variability. The rainfall trends and variations were analyzed using Mann Kendall trend test while farmers’ response was analyzed using SPSS. Contrary to the null hypothesis, the study found a significant trend in rainfall for the last 30 years especially in the MAM seasonal rainfall. The major impact of rainfall variability was low crop yields as reported by 84.6% of all the respondents. The major adaptation strategies practiced by farmers include; soil and water conservation methods, change of planting dates and increased use of organic manure, whereas more than half of the respondents were not practicing any adaptation options. The study recommends that agricultural extension services for example trainings about modern farming practices and climate smart agriculture should be provided to the farmers in order for them to improve and implement better adaptation options to lessen the impacts of rainfall variability. This study will also be of assistance to better inform farmers as well as agricultural policy makers in studying the variability and trends of rainfall in order to characterize the climatology of the area. Keywords: Rainfall Variability, Rainfall Trend, Crop Farmers, Kiboga

We live in a technological era with an increase of information accessible through hard copy or soft copy and there are over 18000 visually impaired children below the age of thirteen in Uganda alone. Most of these children do not have access to reading material like the non-visually impaired and most material available don‟t have braille translated versions. Our project aims at easing access to already available information in different languages and various formats by translating it to grade 1 and grade 2 braille. The purpose of this project is to develop a web application, Text-Braille Converter that converts English documents to grade 1 and grade 2 braille, native language documents to grade 1 braille and audios to grade 1 and grade 2 braille. Overall, this project study provides the visually impaired community with more diverse ways of converting already available information in various formats to a braille format they can utilize.

Introduction: Malaria infection during pregnancy has remained a major public health concern in tropical and subtropical countries with notable risk for the pregnant woman and her fetus. According to the estimated yearly report, the number of pregnant women who were at risk of malaria was about 25 million. It has been reported that in sub-Saharan Africa malaria can cause as many as 10,000 cases of malaria-related deaths in pregnancy per year, usually due to severe maternal anemia. More so, each year, malaria in pregnancy is responsible for 20% of stillbirths and 11% of all newborn deaths in sub-Saharan Africa. World Health Organization, (2018) Uganda is among the ten countries in sub-Saharan Africa that contribute to approximately 70% of global malaria cases and deaths. Uganda is a malaria endemic country with active transmission in 99% of the country. Whereas all people in Uganda are at risk of contracting malaria, children under-5 years of age and pregnant women are the most vulnerable group. MOH- Uganda malaria annual report, (July 2017- June 2018) Objectives: The main objective of this research was to determine the prevalence of malaria in pregnant women attending antenatal clinic at Nsinze Health Centre IV, Namutumba district. Methodology: The study design was cross sectional laboratory and descriptive and the objectives were to determine the prevalence of malaria in pregnant women and to identify the most prevalent species of plasmodium. Approximately 230 women were tested for malaria using both RDT and Microscopy from which the species of malaria parasite were identified and a drug susceptibility test of the plasmodium was carried out to determine the minimum inhibitory concentration of the parasite. The data was analysed by Statistical Package for Social Science (SPSS) and presented as graphs, tables and percentages. Consent was sought from the pregnant women before they participate in the study.

This research analyzed the integration of World Bank Environmental and Social Standards (WBESS) into UNRA’s planning and implementation of road infrastructure projects. The research investigated three research issues, including; the level of integration of WBESS in planning and implementation of road infrastructure projects; factors affecting the integration; and mechanisms for addressing the challenges encountered by UNRA during the integration. Data was collected from 10 respondents drawn from UNRA, World Bank Uganda Country Office, a Supervision Consultant and a Contractor through phone interviews. The phone recordings were transcribed and analyzed using thematic analysis. The following major themes demonstrate the level of integration; undertaking of Environmental and Social Impact Assessments (ESIA), execution of Environmental and Social Management Plans (ESMP), integration of E&S issues in Contract obligations, Project management Restructuring, Project packaging, undertaking Resettlement Action Planning (RAP), E&S compliance monitoring, E&S Compliance Reporting, Stakeholder Engagement Planning, and assessment of Human Resource technical capacity. The following major themes stood out prominently to demonstrate the factors affecting the integration of WBESS; rate of adoption of WBESS by UNRA, adequate stakeholder engagements, enhancement of capacity for implementation of WBESS, conducive working environment for contractors’ workers, workers’ welfare, complexity of WBESS, contractors’ complacency, limited capacity in the Market, land acquisition concerns, strict adherence to WBESS, unempowered contractors’ staff, among others. The possible mechanisms proposed by the respondents to address the challenges encountered by UNRA during the integration process include; need for capacity building on WBESS, embedding of E&S standards in contracts & a cost attachment, continuous training and capacity building for the contractors, enhanced contract/ project management, consistency in application of WBESS, embedding E&S criteria in the procurement process, among others. An important conclusion drawn from this research is that UNRA has the will and determination to execute its road infrastructure projects, based on best available practices that promote environmental and social sustainability. The World Bank one of UNRA’s leading funders of road infrastructure projects, continues to extend development loans and grants to the Authority, to meet its national mandate of developing and maintaining the national road network. However, the Bank should deliberately execute a comprehensive training plan to enhance UNRA’s technical capacity to apply the new World Bank Environmental and Social Framework (WBESF) for its upcoming road infrastructure projects.

Tuberculosis is a contagious disease caused by Mycobacterium Tuberculosis and is one of the leading causes of death worldwide, especially in third world countries such as Uganda. There are several ways to diagnose Tuberculosis but of these, sputum smear microscopy is the commonest method practised in low and medium income areas. However, this method can be error prone and also requires trained medical personnel who are not always readily available because of their few numbers in developing countries. It has therefore become apparent that automating the diagnosis process would go a long a way in alleviating this problem. In this research, we use two pre-trained Convolutional Neural Networks, VGGNet and GoogLeNet Inception v3, to diagnose tuberculosis from Ziehl-Neelsen stained sputum smear microscopic images. These networks are used in three di↵erent scenarios, namely, fast feature extraction without data augmentation, fast feature extraction with data augmentation and fine-tuning. A dataset of 148 images from multiple sources and with di↵erent backgrounds was used. VGGNet achieved accuracies of 76.7%, 80% and 79.6% for the 3 scenarios while Inception v3 achieved accuracies of 86.7%, 77.2% and 76.8% respectively. Additionally, in all three scenarios, Inception v3 computed faster than VGGNet.

Background: The Covid-19 pandemic is argued to have originated from a sea food market in Wuhan city, Hubei Province of China, Due to the contagious nature of the Covid-19, people had to social distance to contain the spread of this novel virus. In order to achieve this measure of social distancing, schools were closed so that children are well distributed among the very many household in Uganda. This school closure affected over 15million learners in Uganda, in order to ensure that there is continuity in their learning process, remote learning was adopted. This meant that learners had to study from home using platforms like TVs, radios and e-platforms like zoom. This study was to analyze the determinants of the adoption of remote learning under Covid. Results: At 5% level of significance, the age of the household head, household size, gender of the household head, regional location of the household, residence of the household (rural/Urban), Nutrition status of the household’s meals, fear for Covid-19 and Education level of the household head were revealed to be significant determinants of the adoption of remote learning during the COVID-19 lockdown period. Conclusion: The analyses on the effect of region and residence on the adoption of remote learning, it was discovered that the rate at which remote learning was adopted across regions and residences (rural/Urban) was not the same. This could be attributed to inequalities in access to electricity and internet. So I recommend that there should be efforts to ensure that the percentage of people connected to the national electricity by reducing the cost of electricity especially during this period where it is a necessity for education access by learners. Different remote learning delivery channels should also be adopted in accordance to the characteristics of the area in which the leaners accessing education from. For regions where the electricity distribution is low, small groups of learners should access tutors their teachers to guide them.

A refractance window dryer of a throughput capacity of 14.5kg/hr was developed to effectively dry food product of 3mm thickness on the conveyor belt. The dryer comprises of a moving conveyor belt onto which food product is placed and drying is achieved by circulating hot water in a tray below the moving conveyor belt. The dryer needed to be automated for effective performance, therefore the research aimed at automating the water circulation system and conveyor belt movement as well as calibrating the dryer for different food products. All systems are designed for quick response in command processing and timely output execution. The automation system comprises of an ARDUINO centered control system, an arrangement of sensors and the output components. The microcontroller receives signals from sensors and sends commands to the respective devices to operate accordingly. A computer program was written in Arduino environment and uploaded on to the control board to process all commands and respond to the appropriate output device accordingly. The program compilation was successful. All systems were first simulated in Proteus simulation environment and the simulation was a success. Components were then tested on onto a bread board before attaching the actual components onto a circuit board. In the design of the food conveyance system to achieve a smooth drive of the belt, a 2.5hp Altivar 31 variable frequency drive was selected which was able to convert single phase to 3phase power to drive the conveyor motor. The VFD mounting volume was designed to be 0.018m3 with a clearance of 50mm on either side of the drive to allow sufficient air circulation around the drive for effective cooling. For effective calibration of the dryer, a 0-5V potentiometer was used for selecting a desired drying temperature. A calibration equation was formulated with the potentiometer maximum angular sweep found to be 295° and was able to attain a 65°C minimum and 94.9996°C maximum system temperature for any selected product. In this research, the system was calibrated for mango, tomato and pineapple but the system can be calibrated for other food products since the optimum drying temperature for most food products falls within the considered range. The calibration system was found to be 97.6% accurate at attaining the desired temperature. All systems were tested for performance and the results showed that the systems run efficiently as simulated. The tests revealed a timely command execution where the system turned on the conveyor belt and water pump only when the desired water temperature set by the knob was reached in the boiler. A product performance tests was done on mango slices at 85°C that revealed a total drying time of 90 minutes to change moisture content from 77.9% to 22.4%. The highest recorded temperature at the tray entrance was 89.9°C after 50 minutes of drying, and the lowest temperature was 31.1°C at the exit of the tray. However, dried slices were found stacked on the conveyor belt after cooling making it difficult to remove. Eliminating the cooling section of the dryer is recommendation to possibly ease dried product removal from the belt based on the previous trials. The food system, water system and conveyor system were found well synchronised in a timely manner. Therefore the setup can be up scaled to a bigger dryer.

Mulago hospital receives about 80% of all casualties due to boda boda accidents in and around Kampala. This amounts to about 260 casualties being brought in on a monthly basis according to Mulago hospital’s Accident and Casualty Ward’s records. This figure indicates that at least 8 accidents happen every day. According to eyewitnesses of these accidents (mainly boda boda riders, even casualties), many people die due to the lack of timely medical attention. Casualties of these accidents often lose their lives because the emergency response provided doesn’t arrive on time. The emergency response, often provided by police stations around the scene of the accident, doesn’t get information about the accident quickly enough to react in a reasonable amount of time. The Crash Detection Helmet with Automatic Emergency Alert comes as a solution to the long response times by emergency alert towards accidents around the city. The helmet will be designed to detect that the wearer has got into a severe accident that requires quick medical attention. Then, notifications will be sent to both the nearest hospital and police station informing them that an accident has happened. The details of that notification include the location of the accident to allow the emergency response to get there quickly.