Two decades ago, African leaders signed the landmark Abuja Declaration pledging to reduce malaria deaths on the continent by 50 percent over a 10-year period marked by 2010. Robust political commitment, together with innovations in new tools and a steep increase in funding, catalysed an unprecedented period of success in global malaria control. According to the World Health Organisation (WHO), 1.5 billion malaria cases and 7.6 million deaths have been averted since 2000.
In the same effort to further reduce malaria incidences globally, in September 2019, the WHO Director-General, Dr Tedros Adhanom Ghebreyesus issued a “malaria challenge,” calling on the global health community to ramp up investment in the research and development of new malaria-fighting tools and approaches. This message was further reinforced in the April 2020 report of the WHO Strategic advisory group on malaria eradication.
Conscious of the above and walking the new Education 5.0 Model launched by the Zimbabwean Government, National University of Science and Technology (NUST) in Zimbabwe Computer Science Student, Bongani Jubani recently developed an application or software to help in the automation of malaria diagnosis process which is expected to boost both the country and Southern Africa’s malaria response strategy.
Education 5.0 Model, which ensures production of goods and services, has added two more areas of focus, Innovation and Industrialisation to Teaching, Research and Outreach. It deviates from the 3.0 Model, which was made up of three core areas — teaching, research and outreach, was inherited from a colonial system which was structured to produce a pool of labourers to service the settler-economy.
NUST Projects Coordinator in Computer Security, Cognitive Science and Digital Forensics, Daniel Musundire said the application system uses artificial intelligence (AI) to detect the presence of malaria parasites in the blood cell of a patient, by just submitting a picture of the patient’s blood cell.
“The developed prototype seeks to use Artificial Intelligence for malaria diagnosis. Firstly the system uses the input of symptoms from the user to predict whether the person has malaria or not. This uses an intelligent agent that calculates the probability of malaria from a set of given symptoms. The second part of the system allows for the upload of microscope images of blood smear. The system classifies whether the cells are infected or not.
“This should assist in environments that are resource constrained. In Zimbabwe, microscopic detection and Rapid Antigen detection are two laboratory diagnosis methods that are used. According to Lancet Clinical Laboratories in Zimbabwe, microscopic detection remains the “gold standard” for the diagnosis of malaria, but is dependent on the experience of the microscopy operator and is also time consuming. At least three smears at different time intervals should be submitted before the diagnosis can be excluded. Microscopic images are not waiting for an expert to read but computers may be used to process multiple diagnoses at a time.”
He said AI in healthcare is an overarching term used to describe the utilization of machine-learning algorithms and software to emulate human cognition in the analysis, interpretation, and comprehension of complicated medical and healthcare data.
In the very complex world of healthcare, he said, AI tools can support health providers to provide faster service, diagnose issues and analyse data to identify trends or genetic information that would predispose someone to a particular disease. When saving minutes can mean saving lives, AI and machine learning can be transformative not only for healthcare but for every single patient.
He said the test run for prototype which is first of its kind in Africa will be made in the second or third quarter of 2021.
Jubani said the system in comparison to contemporary malaria detection methods is faster.
“Remember if diagnosis is delayed it means late treatment and late treatment means loss of life. This system bridge that gap and the results are provided instantly. This system is more accurate compared to other methods currently in use. The system uses up to date technology to improve the diagnosis of malaria. Remember wrong diagnosis means wrong treatment which results in dying “confidently”.
The application is less expensive to use. Drawing the point home, in most Southern African countries, we all know that we have long queues at hospitals where people will be just waiting for the doctor, but now if we have a system that mimic the expert and expensive doctor it means reduction in costs and for example the Cyclone Idai (March 2019) which left many lives at risk of malaria and it’s expensive to deploy doctors in such a high malaria outbreak zone but it’s easy to just deploy this application there and be used by people with no or with minimum expertise,” he said.
Such technological innovations are welcome in the face of worrying World Health Organisation (WHO) malaria statistics, amid revelations that 79 percent of Zimbabweans are at risk of contracting the disease that has claimed more than 226 lives in the country this (2020) year alone.
According to District Health Information Software 2 data approximately 310,000 cases of malaria were recorded in 2019 which is 22 cases per 1000 population. This represented 19 percent increase in number of cases reported in 2018 (approximately 260,000) in Zimbabwe.
According to WHO World malaria report 2020, global malaria mortality fell by 60 percent over the period 2000 to 2019. The African Region achieved impressive reductions in its annual malaria death toll – from 680,000 in 2000 to 384,000 in 2019.
🗳 Since democracy is crucial in a fact-based optimistic world… we remind our readers of the democratic status of the countries we write about:
Zimbabwe has a Global Freedom Score of 29 and has the status Partly Free.
