Automated detection of malaria in Giemsa-stained thin blood smears

Abstract

The current gold standard of malaria diagnosis is the manual, microscopy-based analysis of Giemsa-stained blood smears, which is a time-consuming process requiring skilled technicians. This paper presents an algorithm that identifies and counts red blood cells (RBCs) as well as stained parasites in order to perform a parasitaemia calculation. Morphological operations and histogram-based thresholding are used to extract the red blood cells. Boundary curvature calculations and Delaunay triangulation are used to split clumped red blood cells. The stained parasites are classified using a Bayesian classifier with their RGB pixel values as features. The results show 98.5% sensitivity and 97.2% specificity for detecting infected red blood cells.

Development of inexpensive blood imaging systems: where are we now?

“…Algorithmic, computer-based detection and diagnosis of malaria is the subject of continuing research [41][42][43][44][45], and could aid in the implementation of these new technologies by lowering barriers to adoption and increasing access to care. However, an intriguing concept has been recently developed that combines the power of digital imaging with human ingenuity.…”

Malaria Parasitaemia Estimation � A new perspective (Preprint)

“…Author Suryawanshi et al [39] performed image binarization by using Poisson distribution based minimum error threshold value. Authors Mushabe et al [40], Damahe et al [41] used Zack thresholding method on the 'V' or 'value' component of HSV image for segmentation. Authors Purwar et al [42] have used Active Contour model for segmentation of red blood cells.…”

“…The authors Nugroho et al [24] used K-NN classifier with 'S component' in HSV colour space for segmentation. Mushabe et al [40] used K-NN classifier to identify the parasite and non-parasite regions. Nasir et al [48], used the saturation channel of image in HSI colour space and moving Knearest neighbour (MKNN) to differentiate parasite containing red blood cell.…”

Automated system for characterization and classification of malaria‐infected stages using light microscopic images of thin blood smears

“…Recently RGB pixel information‐based malaria diagnosis using Bayesian classifier was reported in Mushabe et al . (). Based on this extensive literature survey, it is observed that so far the maximum number of microscopic features (= 83) is considered by Tek et al .…”

“…Circle-ellipse fitting search algorithmbased malaria diagnosis was reported recently (Sheikhhosseini et al, 2013). Recently RGB pixel information-based malaria diagnosis using Bayesian classifier was reported in Mushabe et al (2013). Based on this extensive literature survey, it is observed that so far the maximum number of microscopic features (= 83) is considered by Tek et al (2010) for malaria detection based on lesser sample size.…”