Application of Artificial Intelligence for quantifying Plasmodium berghei in blood samples from infected mice.

Abstract

Background objectives: In malaria infection, quantifying blood parasitemia is a critical step for evaluating the severity of the disease. This has generally been conducted manually, and thus, its accuracy depends on the expertise of technicians. There is an urgent need for an automated technique to overcome manual errors. The aim of the study is to find an alternative method for counting malaria blood parasitemia.

Methods: This study evaluated the accuracy of automated counting using QuPath compared to manual counting. GraphPad Kappa evaluated agreement between high and low parasitemia in both counting methods using Cohen's test.

Results: QuPath was revealed to be a promising method that has fair agreement and no statistically significant differences compared to manual counting.

Interpretation conclusion: In conclusion, automated quantification is suggested to be a time- and effort-saving technique and, therefore, a worthwhile alternative to manual counting.