Plasmodium Parasite Detection on Thin Blood Smear Image using Double Thresholding and BLOB Analysis

Author(s):  
Jullend Gatc ◽  
Febri Maspiyanti
IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 93782-93792 ◽  
Author(s):  
Muhammad Umer ◽  
Saima Sadiq ◽  
Muhammad Ahmad ◽  
Saleem Ullah ◽  
Gyu Sang Choi ◽  
...  

Author(s):  
P.A Pattanaik ◽  
Tripti Swarnkar

The genus Plasmodium parasite causes malaria infection. Fast detection and accurate diagnosis of infected and non-infected malaria erythrocytes from microscopic blood smear images open the door to effective assistance and patient-specific treatment. This article presents a comparative experimental analysis of visual detection of infected erythrocytes malaria parasites via the most efficient morphological techniques from gold standard blood smear images. In this article, twelve different widely-used morphological algorithms are evaluated followed by a random forest classifier for detecting infected erythrocytes based on their performance vis-a-vis microscopic blood smear images. Accurate detection of infected malaria erythrocytes is done using the two ranges of blood smear image datasets with varying malaria parasite density. Finally, compared to 11 morphological techniques in terms of accuracy, sensitivity, and specificity, the qualitative assessment of experimental results unveil that the Histogram method offers more meaningful and impactful findings.


Sign in / Sign up

Export Citation Format

Share Document