Identifying Plasmodium Parasites in Blood Smears with Deep Learning Techniques

Abstract

Malaria is a life-threatening disease caused by Plasmodium parasites that are transmitted to people through the bites of infected female Anopheles mosquitoes and continues to be a leading public health challenge worldwide. Five species of Plasmodium infect humans (with P. falciparum being the most dangerous, which can risk severe complications such as organ failure and death). P. Vivax and p. Infections with ovale, although usually less severe, can recur from a dormant parasite in the liver. Globally, in 2018, more than 400,000 deaths occurred for malaria, mainly p. Due to Phalsiparum, there were cases of travel or local transmission (WHO, 2017; WHO, 2019), along with local regions in Africa as well as non-sustainable regions. It requires quick diagnosis with blood smear microscopy, which is the standard of gold to determine the identity of parasites and species. Although rapid clinical trials (RDTs) are widely available, they are often low in sensitivity, while polymerase chain reaction (PCR) provides desired sensitivity but impractical to immediate diagnosis.Severe malaria poses specific problems, especially in inexperienced individuals and pregnant ladies, typically related to impaired consciousness and metabolic acidosis, which correlate with worse outcome. Timely diagnosis and treatment are crucial in controlling malaria, which also depends on continuing strategies to cope with the disease's burden, especially in susceptible populations such as pregnant women and non-immune travelers and to provides rapid and accurate detection of Plasmodium parasites on an app that interprets images of blood smears. These method is enabling quick medical treatment, which leads to better results in the treatment of malaria.

Topics & Concepts