Computer vision for microscopy diagnosis of malaria

Abstract

This paper reviews computer vision and image analysis studies aiming at automated diagnosis or screening of malaria infection in microscope images of thin blood film smears. Existing works interpret the diagnosis problem differently or propose partial solutions to the problem. A critique of these works is furnished. In addition, a general pattern recognition framework to perform diagnosis, which includes image acquisition, pre-processing, segmentation, and pattern classification components, is described. The open problems are addressed and a perspective of the future work for realization of automated microscopy diagnosis of malaria is provided.

Figure 1

Examples of stained objects. (a, b) white blood cells, (c, d) platelets, (e)-(h) artefacts, (i)-(l) P. falciparum ring, trophozoite, gametocyte, schizont, (m, n) P. malariae ring and schizont (o, p) P. ovale and P. vivax trophozoites, (q, r) P. vivax ring and gametocyte, (s) P. vivax ring, (t) extracted stained pixel group. (S, green region(s)) and the stained object (S_b, red region including the green one).

Figure 2

Stained object classes: in a Giemsa-stained blood film an observed stained object can be a parasite from one of the four species of Plasmodium or a regular blood component such as white blood cell, platelet. Artefact class represents bacteria, spores, crystallized stain chemicals, particles due to dirt, RBC anomalies (e.g. Howell-Jolly bodies, iron deficiency, reticulocytes), and other peripheral blood parasites.

Figure 3

Examples of Giemsa-stained (a) thin and (b) thick blood film smear images, (c) a concentrated (thick) field of a thin blood film smear.

Figure 4

Size granulometry vs. area granulometry. (a) negative image of the grey level sickle cell image, (b) granulometry using disk shaped structuring elements, (c) area granulometry, (d) area granulometry based cell size estimation varies in different fields of a thin film although the magnification is constant.