Inventors: Oliver Higgins, Jonathan Cooper, Julien Reboud; University of Glasgow
Schistosomiasis (also known as bilharzia) is a disease caused by infection with parasitic flatworms called schistosoma. The disease is primarily transmitted through contact with contaminated water and is common in areas with poor sanitation facilities. It causes a range of chronic symptoms such as liver damage, kidney failure and bladder cancer. Aside from increased mortality, the disease causes insidious developmental effects, with anemia, stunted growth, and impaired cognitive development prevalent among infected children. The disease is widespread in Africa, Asia and South America, affecting an estimated 252 million people worldwide in 2015, and is second only to malaria in its economic impact in tropical countries. Despite this, it is classed as a neglected tropical disease.
Diagnosis of the disease is performed by manual inspection of faecal or urinary samples to identify the parasite’s eggs. Currently this process is carried out manually using laboratory-based microscopes and can take 10-15 minutes of work per sample. The process is time consuming, arduous and is associated with a significant personnel cost due to the highly skilled nature of the work. Because of the difficulty of performing diagnosis at scale, many countries have adopted an approach of mass drug administration in areas where the disease is known to be present. However, the speed of reinfection with the parasite means that this process needs to be repeated regularly to be effective and there is evidence that the policy is starting to lead to drug-resistant strains of the parasite.
The team at Glasgow is therefore working on a smartphone-based diagnosis system, to automate the diagnosis of schistosomiasis. The device uses a low-cost 3D-printable microscope stage to scan patients’ samples, along with a deep learning-based decision support application on the smartphone, which automatically identifies the presence of eggs. The aim is to harness the power of mass market hardware to provide inexpensive, battery-powered diagnostics which can be more easily transported to, and used, in areas with schistosomiasis outbreaks.
The automated decision support system reduces the requirement for manual sample processing and documentation, and allows a digital record of the test to be catalogued in the cloud. Furthermore, the location data of tests can be logged in order to identify epidemiological trends so that drug administrations campaigns can be targeted more effectively.
Early field trials have been carried out in Uganda, and so the challenge for the i-Team is to investigate where else the system should be trialled for maximum impact, and where there might be sources of funding available to support the rollout of the system.