Beijuka Bruno
Research Engineer
Research Interests
AI in Health, Machine Learning in Low resource settings
Projects
Title: Tuberculosis in households with infectious cases in Kampala city: Harnessing health data science for new insights on TB transmission and treatment response
The DS-I Africa-TB project is a multidisciplinary effort focused on advancing the understanding of tuberculosis transmission and treatment responses within households in Kampala City. By harnessing health data science and artificial intelligence (AI), the project seeks to uncover critical insights that could inform TB control strategies and improve patient outcomes.
Title: Exploring the Role and Feasibility of Natural Language Processing Techniques to Improve Mental Health Services in Uganda and Tanzania
This project focuses on developing natural language processing (NLP) and automatic speech recognition (ASR) tools for mental health services in Uganda and Tanzania. Using call center conversations in English, Luganda, and Swahili, the team is creating datasets and models for transcription, translation, emotion and sentiment analysis, and conversation quality assessment. By advancing NLP for under-resourced African languages, the project aims to enhance digital mental health services and improve access to care in low-resource settings.
Title: Automatic Speech Recognition for African Languages: How Much Data is Enough?
This project investigates how much speech data is required to build accurate automatic speech recognition (ASR) systems for low-resource African languages. By leveraging open datasets like Mozilla Common Voice and developing benchmark corpora across African languages, the study evaluates different ASR models (supervised, self-supervised, and transfer learning) in both general and domain-specific contexts, such as health, education, and agriculture. The findings will inform sustainable data collection strategies and advance open-source ASR tools for African languages.
Title: Computer-Aided Diagnosis System for Cervical Cancer Screening Using Mobile Colposcopy Images
This project develops an AI-assisted system for early detection of cervical cancer using mobile colposcopy images. By leveraging machine learning and image analysis, the system aims to support clinicians in low-resource settings where specialist expertise and diagnostic infrastructure are limited. The project is funded by the Makerere Research and Innovations Fund and SPIDER, in collaboration with the Uganda Cancer Institute and the Science, Technology, and Innovation Secretariat (STI).
Previous Education and Professional Experiences
Education
BSc. Computer Engineering - Makerere University, 2021-May, 2024
Experience
Research Engineer - Marconi Lab (June 2024 - to date)
Machine Learning Intern - netLabs!UG (June 2023 – August 2023)
Backend Engineer Trainee - Crane Cloud (June 2022 - December 2022)