ABOUT THE MILLENNIUM FELLOWSHIP - CLASS OF 2024
United Nations Academic Impact and MCN are proud to partner on the Millennium Fellowship. This year, 52,000+ young leaders applied to join the Class of 2024 on 6,000+ campuses across 170 nations. 280+ campuses worldwide (just 5%) were selected to host the 4,000+ Millennium Fellows.

UNITED NATIONS ACADEMIC IMPACT AND MCN PROUDLY PRESENT ALABYEKKUBO SSUUBI BRIAN, A MILLENNIUM FELLOW FOR THE CLASS OF 2024.
Makerere University | Kampala, Uganda | Advancing SDG 17, SDG 4, SDG 9 & UNAI 1

" I am excited to be a Millennium Fellow because it provides a unique opportunity to collaborate with like-minded individuals who are passionate about creating meaningful change. I am driven by the belief that technology and innovation can transform communities, and through this fellowship, I aim to amplify my impact, particularly in using AI, data science, and software development to solve pressing challenges in health and education. "
Millennium Fellowship Project: Natural Language Explanations For Malaria Microscopy
Malaria is a infection caused by Plasmodium which is a protozoan transmitted by various types of mosquitoes including theAnopheles type of mosquitoes. There are mainly four classes of Plasmodium, namely P. vivax, P. malariae, P. ovale and P. falciparum. This life-deadly disease is among the major world public health problems, which causes over 450 million infections all around the word, and relatively one million deaths each year. It not only causes illness and suffering but also traps families in a cycle of poverty due to the costs of treatment. Currently, a greater half of the world’s population, predominantly residing in sub-Saharan part of Africa, is at risk of contracting this deadly disease and dealing with its financial burdens. We proposed to present a novel approach that can enhance recording and interpretation of malaria microscopy results. Leveraging Visual Transformers(VIT) and GPT-3 technologies, our research introduces a system that provides clear, natural language explanations for malaria diagnosis results obtained through microscopic analysis. By bridging the gap between complex medical imagery and comprehensible human language, this study aimed to improve diagnostic accuracy, empower healthcare professionals, and advance the fight against malaria, a deadly tropical disease.The Vision Transformer helps to analyze and classify microscopy images, leveraging its superior performance in image recognition tasks. Simultaneously, GPT-3 is utilized to generate detailed and coherent natural language explanations based on the ViT’s outputs. Therefore in this research, we focused on addressing critical issues in malaria diagnosis and microscopy, including limited transparency in microscopy and lack of explanations for diagnostic results thus our research aims to contribute and address this problem by providing simple and understandable NLEs (Natural Language Explanations) for microscopic results thus easing the tedious and complex process of interpretation and recording of result, since were mainly focusing on specific malaria and plasmodium contributing factors. It is an essential and important task to accurately diagnose malaria and quickly to ensure effective treatment and proper control of the above mentioned disease. The use of microscopic images is
a crucial tool in diagnosing malaria by examining blood samples to ascertain the presence and count of plasmodium parasitesin the blood sample. However, the current methods and technologies used to face a significant array of challenges, which our research aimed to resolve. The current approaches lacked interpretability, hindering accurate and efficient diagnosis. This research report aimed to address this problem by developing a VIT framework that enhances understanding and trust in malaria microscopy diagnosis results. Our objectives included improving accuracy as well as efficiency of diagnosis, and advancing malaria diagnosis techniques too. Our contributions look to include the development of the natural language explanations framework and its impact on improving interpretability and trustworthiness in malaria microscopy.
About the Millennium Fellow
I am Alabyeekubo Ssuubi Brian, a passionate software developer, machine learning engineer, and DevOps engineer from Bukumula, Mityana, Uganda. With roots in Burundi, I am known for my innovative approach to problem-solving. I am a finalist at Makerere University and have gained significant experience through various roles, including at DFCU Bank. I am committed to leveraging technology to make impactful contributions, particularly in AI and software development to achieve the global SDGs.








