Applicants submitted a proposal to one of two domains: data and analytics or diagnostics. Three applicants from each domain became finalists.
2021 Finalists and Proposals
Data and Analytics
Kareen Atekem - Sightsavers - Cameroon
Proposal: Developing a Trap for African Chrysops Flies to Accelerate Onchocerciasis Elimination in Areas co-endemic with Loiasis
Fourteen million people in central Africa are excluded from mass drug administration (MDA) for onchocerciasis due to co-endemicity with loiasis, including in Cameroon. Severe adverse events (including death) occur when highly loiasis-infected individuals receive ivermectin. Exclusion creates potential reservoirs of transmission and increases health inequalities. Current Chrysops catching methods use human baits, exposing individuals to potentially loiasis-infectious bites. The lack of a Chrysops trap also means the continued use of unethical human landing catches in any programmatic scenario requiring assessment of loiasis. This study aims to develop a trap for Chrysops flies, vectors of loiasis, based on their attraction to horizontally polarized light. Production of multiple trap prototypes and comparing their efficacy (using analysis of trap catch data in a randomized square design), will enable identification of an optimal design. In addition to vector control, the trap will provide a solution to the lack of scalable ethical options for catching Chrysops flies. Watch the final pitch video.
Amber Barton and Martin Holland - London School of Hygiene and Tropical Medicine - United Kingdom
Proposal: Using Artificial Intelligence to Predict Trachoma Progression from Photographs and in vivo Microscopy Images
Trachoma is a neglected tropical disease caused by ocular chlamydia trachomatis infection. Active disease manifests in the form of conjunctival follicles and inflammation, with repeated episodes of disease resulting in scarring, trichiasis and irreversible blindness. Under the WHO simplified system, active trachoma diagnosis is highly subjective and it is currently difficult to predict on the basis of active disease which patients are at risk of scarring. The study aims to use artificial intelligence to objectively classify participants into clinical categories and predict their likelihood of going on to develop scarring and blindness. This is a non-invasive technique allowing high-resolution cellular level imaging of the ocular surface, revealing microscopic features of disease such as changes in connective tissue organization and phagocyte infiltration. Watch the final pitch video.
Kevin McRae-McKee and Atia Abdalla - Oriole Global Health - United Kingdom
Proposal: Development of an Interactive Dashboard to Optimize the Planning of NTD Programs
The use, or lack thereof, of digital tools that facilitate planning, implementation, and reporting is currently a major limitation of NTD programs. While tools of varying quality and flexibility do exist, ministry-managed tools that provide a complete overview/picture of disease epidemiology and partner activity to support effective planning of NTD programs do not. There is an urgent need to develop standardized and, at the same time, flexible digital tools to optimize the use of existing programmatic data in resource-poor settings and to engage policy makers by encouraging evidence-based decision making. This project will aim to develop an easily accessible, no-cost/open-source interactive tool that combines the analysis of complex and often sparse and disjointed data sets with spatial statistics in such a way that policy makers and program managers can interact with their data in real time. Such a tool would help bring government and partners together with a single narrative, while rapidly identifying gaps in program coverage, opportunities for efficiencies in delivery, and supporting an understanding of historical program coverage and future NTD program resource needs and activities. Watch the final pitch video.
Diagnostics
Anthony Ablordey - Noguchi Memorial Institute for Medical Research, University of Ghana - Ghana
Proposal: Development of the Dried Reagent Based IS2404 Loop-Mediated Isothermal Amplification Test for the Diagnosis of Buruli Ulcer Disease
Although an effective cure for Buruli ulcer is available, a major bottleneck for control programs is early laboratory diagnosis. A simple, rapid and sensitive test that can be deployed in endemic settings is needed to in order to control Buruli ulcer. The Dried Reagents-Based IS2404 loop-mediated isothermal amplification (DRB-IS2404 LAMP) test that can be easily deployed and used in BU endemic areas will facilitate early and timely diagnosis and improve patient care considerably. This will allow for early antibiotic treatment, shorten hospitalization times, reduce sequelae like permanent disabilities, minimize surgical interventions and prevent stigma and socio-cultural drawbacks. The study will help determine the feasibility of implementing the DRB-IS2404 LAMP testing for BU diagnosis in district health facilities in BU endemic countries. Watch the final pitch video.
Temitope Agbana - Delft University of Technology - Netherlands
Proposal: Smart and Affordable Digital Diagnostic for Schistosomiasis
Schistosomiasis is a major health burden in humans and livestock. Early diagnosis can prevent significant health damage, but current diagnostic tools are labor intensive, lack quality control or require expensive equipment. The aim of this project is to develop a smart and affordable digital tool for diagnosis of schistosomiasis in rural endemic areas. This tool, the Schistoscope, is a novel methodology that combines technical optics and specialized data-driven algorithms to realize an integrated, portable and reliable smart optical diagnostic system. Generated diagnostic data can be analyzed offline and uploaded onto remote servers. This will accelerate prompt, large-scale diagnosis in community mapping for instituting intervention and monitoring. Watch the final pitch video.
Hugues Nana Djeunga and Guy Sadeu Wafeu - Center for Research on Filariasis and other Tropical Diseases -Cameroon
Proposal: Cell Phone-Based Imaging for the Rapid Point-of-Care Diagnosis for Scabies: a Machine Learning Approach
Scabies is a skin NTD highly endemic in low-income settings; its diagnosis can be done from skin lesions and symptoms features, though it is usually poorly diagnosed by non-specialists. The main critical action is therefore to develop a diagnostic tool for daily practice and mapping to estimate the disease burden. Point-of-care cell phone imaging was considered as cross-cutting technology to improve integrated management of skin NTDs. The project aims to put new technologies at the service of healthcare by developing a cell phone-based application for the rapid diagnosis of scabies at point-of-contact using artificial intelligence. Deep learning will be used to identify scabies lesions using convolution neural network algorithms. Specific questions on symptoms’ features and disease history will be used to improve the accuracy of the diagnosis. The diagnosis will therefore be based both on the automatic recognition of lesion pictures and specific questions to patients. Watch the final pitch video.