2011 International Conference on Image Information Processing (ICIIP 2011) Proceedings of the 2011 International Conference on Image Information Processing (ICIIP 2011) 978-1-61284-861-7/11/$26.00 ©2011 IEEE Plasmodium vivax segmentation using modified fuzzy divergence Madhumala Ghosh, Devkumar Das, Chandan Chakraborty School of Medical Science and Technology Indian Institute of Technology Kharagpur, India madhumala_ghosh@rediffmail.com , Ajoy K Ray Bengal Engineering and Science University Howrah, India akray_2004@yahoo.com Abstract—This paper aims at introducing a new approach to Plasmodium vivax (P. vivax) detection from Leishman stained thin blood film. This scheme follows retrospective study design protocol where patients were selected at random in the clinic. The scheme consists of two main stages – firstly artefacts reduction, and secondly fuzzy divergence based segmentation of P. vivax infected region(s) from erythrocytes. Here, malaria parasite segmentation is done using divergence based threshold selection. Fuzzy approach is chosen to minimize ambiguity inherent in the microscopic images. Divergence algorithm is derived from Cauchy membership function to overcome the drawbacks in comparison with other well known membership functions. Keywords-component; formatting; style; styling; insert (key words) I. INTRODUCTION Malaria, one of the most serious worldwide health problems causes 1.5-2.7 million of deaths per year [1]. The worldwide annual economic trouble of malaria, including spend on prevention and treatment as well as loss in efficiency due to sickness, is estimated at US$800 million in 1995 [1]. In India approximately 30-40% of people are affected by Plasmodium vivax that is reported in National Vector Borne Disease Control Program (NVBDCP) data in the year of 2010. So, fast and accurate diagnosis is required to facilitate prompt treatment to control malaria. In today’s diagnostic paradigm, microscopic imaging technology has immense contributions in generating fruitful medical images, which essentially become the basis for medical experts to make better decisions. In practice, pathologists visualize the abnormalities if any, in the images through microscope based on their knowledge from the view point of intensity, morphology, texture etc. based features. Usually small scale differences in the features are overlooked by human eyes especially for the border-line diagnostic scenario. In order to circumvent this problem, it is more worthwhile to develop computer-assisted automated screening scheme for automatically characterizing the abnormalities, especially in complicated cases where experts fail to take decision. In doing this, microscopic information needs to be analysed quantitatively maintaining the biological integrity in the system. Here an attempt has been made to develop an automated pattern analyser to detect the malaria parasite (Plasmodium vivax) which helps to reduce the time efficiency as well as inter and intra-observer variations. Some works are reported in the literatures. Tek [2] introduces automatic detection of malaria parasite based on color histogram. Diaz et al. [3] shows the Quantification and classification of Plasmodium falciparum infected erythrocyte. Ross et al. [4] used the morphological and novel thresholding selection technique for identify erythrocyte and possible parasite. Makkapati et al. [5] segment the malaria parasite in HSV color space. Raviraja et al. [1] detects the red blood cells that are infected by malaria parasites using statistical based approach. Dempster et al. [6] applies Mathematical morphology and granulometry approach for automatic estimation of parasitemia with no human intervention. Toha et al. [7] segments the malaria parasite using Gray level thresholding. Garcia [8] approaches automatic malaria infected Cell counting. Halim et al. [9] estimates parasitemia based on pattern matching with parameter optimization and cross validation against the expected biological characteristics, red blood cells are determined. This paper aims at introducing an automated approach to parasitemia detection from ‘Leishman’ stained thin blood film infected with P. vivax using fuzzy divergence based thresholding techniques. The proposed scheme is a combination of four different stages: a pre processing step for correcting the noise, a segmentation step which is done by threshold selection using fuzzy divergence, textural feature extraction and classification which identifies normal and infected erythrocytes from their textural features. II. MATERIAL AND METHOD A. Human blood smears preparation Retrospective study design is here followed in selecting the subjects from the population at the Dept. of Hematology, Midnapur Medical College & Hospital and Medipath Laboratory, Midnapur, West Bengal. The blood samples from there malaria infected patients are collected and processed as follows. Thin blood smear is prepared on a clean and disinfected slide and stained with Leishman for visualizing different cellular counterparts. In the laboratory, firstly 5-7 drops Leishman is applied to the slide with the specimen. After 5 minutes, 10-12 drops of a buffer solution (pH 6.8) is added and mixed with the stain, then the specimen is left