Projects
MUDSReH supports two core research projects: i) Multimodal Database of Retinal Images in Africa (MoDRIA) and ii) Automated Visual Evaluation and geospatial Mapping (AVE-Map) for cervical cancer screening optimization in Sub-Saharan Africa. In addition, we plan to expand to new projects in dermatology, radiology and microbiology/pathology
Datasets
MUDSReH collects clinical datasets for posterior segment eye diseases and cervical cancer in a predominantly African setting. With plans for expansion to new projects and sites that will collect additional datasets in dermatology, radiology and microbiology/dermatology. These datasets will be applied to Artificial Intelligence (AI) models to improve health outcomes.
Partners
MUDSReH is a data science hub under Mbarara University of Science and Technology. It collaborates with Massachusetts General Hospital (MGH), Massachusetts Institute of Technology (MIT), Kwame Nkurumah University of Science and Technology (KNUST) and College of Ophthalmologists of Eastern, Central, and Southern Africa (COECSA).
Overall Goal of the Research Hub
To produce a user-friendly data science center of excellence that focuses on the translation of clinical datasets into implemented clinicial interventions in sub Saharan Africa.
Plan of the Research Hub
Medical images offer great potential for supporting clinical diagnosis and care management; however, their utility for African researchers and clinicians are limited by the number of images available, insufficient linkage to relevant clinical data, limited data from African populations for proper image interpretation, and poor implementation in clinical practice. The Mbarara University Data Science Research Hub (MUDSReH) is designed to overcome these barriers through robust Data Management and Analysis and Administrative Cores, initially supporting projects in posterior segment eye disease and cervical cancer with plans for expansion to new projects and sites, as well as training in data science and implementation science.
MUDSReH Core Projects
Multimodal Database of Retinal Images in Africa (MoDRIA)
Posterior segment eye diseases are the second leading cause of blindness in sub- Saharan Africa, resulting in loss of quality of life and economic productivity. Imaging of the back of the eye (retina) can help us diagnose most blinding posterior segment eye diseases (PSED). MoDRIA will create the first publicly accessible retinal imaging database (with metadata) based on the African population. We will develop, test and validate PSED screening algorithms using data from an African population.
Read more
Automated Visual Evaluation and Geospatial Mapping (AVE-Map)
Automated Visual Evaluation and Geospatial Mapping (AVE-Map) is a clinical trial aimed at improving cervical cancer screening and accessibility through use of artificial intelligence and geospatial analysis. We aim to leverage and develop data science expertise at our sites to first optimize and then combine AVE-based screening by health workers at peripheral health facilities with geospatial-analysis and needs-driven assessment to inform scale-up of cervical cancer screening in Ghana and Uganda.
Read more
Data Management and Analysis Core (DMAC)
DMAC is one of the two cores in MUDSReH to support research projects. DMAC's goal is to establish, an Open Data Science Platform (ODSP), the MUDSReH-DI, and support systematic data capture, data integration and harmonization, and hypothesis- and local data-driven research and training, through protecting, preserving, processing, moving, securing, providing and facilitating analysis of health-related data, and also support the people, the processes and the policies to create a health data science ecosystem in an African
Read more
Administrative Core
Administrative Core manages the “MUST Data Science Research Hub (MUDSReH)” through a multiple Principal Investigator (MPI) structure where Prof. Celestino Obua, is the contact PI and overall Hub Director working with Drs. Leo Celi and Jessica Haberer as MPIs. MUDSReH is housed in the Office of Research Administration (ORA) where the core research administration staff have all received NIH-supported training in research administration and management.
Read more
Artificial Intelligence
Artificial intelligence (AI) refers to the simulation of human intelligence in machines that are programmed to think like humans and mimic their actions. The term may also be applied to any machine that exhibits traits associated with a human mind such as learning and problem-solving. The ideal characteristic of artificial intelligence is its ability to rationalize and take actions that have the best chance of achieving a specific goal. We apply AI techniques to medical images in MUDSReH.
Read more
Internet Of Things
The Internet of Things (IoT) is a network of physical objects which sense, and transfer data over the internet without human intervention. Examples of IoT include driver-less cars, smart homes, wearables, smart cities, smart door locks, air quality sensors, etc. MUDSReH leverages the power of IoT to translate medical images into similar language which can be understood by health workers at any level. We plan to IoT to improve disease diagnosis in LMICs.
Read moreMUDSReH Pipeline
