According to a paper published in BMC Bioinformatics earlier this week, researchers are utilising a machine learning (ML) tool to speed up the turnaround time for research on malaria. ML is a subset of artificial intelligence (AI).
The goal of machine learning models is to generate precise predictions or judgements based on previous data. A series of algorithms create models based on sample data, often known as training data.
Researchers from the University of Glasgow in Scotland and the Ifakara Health Institute in Tanzania claim that understanding a mosquito’s capacity to spread malaria, a fatal disease conveyed by infected Anopheles mosquitoes, is made possible by learning how old the insect is. They did add, though, that the current methods for foretelling this are expensive, labor-intensive, and frequently prone to human error.
In order to determine the biochemical makeup of the mosquitoes, this study focused on strains of mosquitoes that were grown in laboratories in the two study locations, Tanzania and Scotland. After that, they trained models to predict mosquito age using ML approaches. Studying these traits is crucial in the fight against malaria because, for instance, older mosquitoes are more likely to transmit malaria than younger ones, but mosquitoes that prefer to feed on people are more likely to transmit malaria than those that prefer other animals.
Machine learning is a more effective choice for forecasting mosquito ages than the current technologies, which are time-consuming and expensive, according to Emmanuel Mwanga, the study’s lead author and a research scientist at Ifakara Health Institute.
The study’s results demonstrate that machine learning models may be used to estimate the ages of mosquitoes from various populations with an accuracy of roughly 98% for predictions of the same mosquito ages. However, the technique can assist entomologists in shortening the time and labour needed to dissect a sizable number of mosquitoes.
According to Mwanga, using ML approaches could “save time and resources that can be used for other areas of malaria control and elimination activities.” Accurately predicting these parameters “can assist identify high-risk people and focus interventions more effectively.”
In the end, he said, “this can result in a decrease in the number of malaria cases and deaths, which is a crucial step toward attaining zero malaria.”
The results, according to researchers, are important for policymakers because they will simplify resource allocation, help identify trends, and support the creation of effective strategic plans for the eradication of this disease.
The researchers emphasised that additional investigation was necessary because the study mainly focused on Anopheles arabiensis, a particular species of mosquito found in just two nations. In order to improve malaria interventions, he added, malaria scientists must better understand the precise age, host preferences, and species of the malaria-carrying agents. “It’s vital to test the findings on mosquitoes from diverse places and species,” he said.
The World Health Organization estimates that in 2021, there will be 247 million cases of malaria worldwide, with about 95% of those cases occurring in the African continent. In India, malaria is still a significant public health issue. The nation is responsible for 2-3% of the world’s malaria cases and 2% of the disease’s fatalities (52% of all malaria deaths outside of sub-Saharan Africa). In India, the north-eastern, eastern, and central regions are particularly affected by the deadly disease.
The key to eradicating malaria, according to scientists, is the adoption of cutting-edge methods for managing mosquito populations and halting the disease’s spread.