A Novel Algorithm for Segmentation of Parasites in Thin Blood Smears From Microscopy Using Type II Fuzzy Sets and Inverse Gaussian Gradient

A Novel Algorithm for Segmentation of Parasites in Thin Blood Smears From Microscopy Using Type II Fuzzy Sets and Inverse Gaussian Gradient

Madhu Golla
Copyright: © 2019 |Pages: 22
DOI: 10.4018/IJCVIP.2019070101
(Individual Articles)
No Current Special Offers


Malaria is the leading health problem globally in the equatorial regions. Even after campaigning and control by the World Health Organization (WHO), malaria is a high mortality-causing infectious disease due to inappropriate and late diagnosis. To prevent getting affected by malaria, the diagnosis should be in the early stage and accurate. This research presents an innovative approach for erythrocyte segmentation and malaria parasite detection from microscopic images. In addition, the author utilizes Einstein t-conorm function to generate new fuzzy membership function which is used to segment the infected regions of the blood cells in microscopic images. Finally, the inverse Gaussian gradient function is used for the final segmentation. This technique is compared with state-of-the-art segmentation techniques such as k-means, fuzzy c-means segmentation, and spatial intuitionistic fuzzy c-means (SIFCM) segmentation approaches. Experimental results are validated qualitatively, and it is deduced that the proposed method is the robust segmentation method for malaria parasite detection.
Article Preview

1. Introduction

Malaria is a severe tropical infectious disease caused by the female anopheles’ mosquito and that is infected by Plasmodium species(Cojoc,2012). In 2017, World Health Organization (WHO) report, Malaria treated as one of the major health problems, it’s affects approximately 698 million people around the world (WHO, 2017). In India the total number of malaria cases were reported around 1.31 million in the global population, 6% of deaths, and 51% of the Plasmodium. vivax (WHO,2017) (Malaria in India, 2017). In practice, the pathologists are visually examining the blood smears through microscopy for the diagnosis of malaria. Still, this type of approaches are subjective, time-consuming, and late diagnosis (Loddo, 2018). To address these issues, a fast and accurate computational diagnosis is required a suitable intervention, especially in remote areas with limited healthcare services. However, in microscopic images, the malaria parasite presents in the form of the nucleus and cytoplasm, which is not an easy task to detect and segment the infected region of the parasite in microscopic blood images. To segment these regions, need segmentation process to split image cells into different homogeneous groups with similar features of brightness, color, contrast, and gray levels (Sharma and Aggarwal, 2010) (Buenestado, 2018) (Wuli Wang, 2017)]. In general, segmentation is to transform an image into different segments by extracting important pixels and designate a marker to every pixel of the image (Buenestado, 2018). However, Image segmentation is a major phase in microscopic imaging analysis and, it plays important role in digital image analysis for medical diagnosis process, which is used for cell detection, recognition, and model representation (Kun He, 2016) (Narjes Ghane, 2017). Still, there is no common approach for medical image segmentation, because every image scheme has its own precise constraints (Sharma and Aggarwal, 2010).

In view of the above issues and challenges, this research developed a innovative approach for segmentation of parasites from microscopic blood images in the diagnosis of malaria, which address the following significant features:

  • To transform the microscopic images into a grayscale image and performed non-local means denoising technique on this grayscale image to reduce the noise levels.

  • To calculate the Gaussian membership function on this denoised images by using lower membership and upper membership functions.

  • By using Einstein t-conorm, developed a new membership function to segment and highlight the infected parasite cells from microscopic images.

  • Finally, Inverse Gaussian gradient function is applied on above categorised image, which obtain the final segmentation of malaria parasites.

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 13: 1 Issue (2023)
Volume 12: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 2 Issues (2016)
Volume 5: 2 Issues (2015)
Volume 4: 2 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing