An Intelligent Recognition Procedure for Trophozoite Stages of Plasmodium Knowlesi Malaria

Authors

  • S.N.A.M. Kanafiah School of Mechatronic Engineering, University of Malaysia Perlis, Perlis, Malaysia.
  • M.Y. Mashor School of Mechatronic Engineering, University of Malaysia Perlis, Perlis, Malaysia.
  • Z. Mohamed Department of Microbiology and Parasitology, University of Sains Malaysia.
  • J.X. Teh School of Mechatronic Engineering, University of Malaysia Perlis, Perlis, Malaysia.
  • S.A. Abdul Shukor School of Mechatronic Engineering, University of Malaysia Perlis, Perlis, Malaysia.
  • N.Z. Zakaria Department of Microbiology and Parasitology, University of Sains Malaysia.
  • H.N. Lim School of Mechatronic Engineering, University of Malaysia Perlis, Perlis, Malaysia.
  • H. Ali School of Mechatronic Engineering, University of Malaysia Perlis, Perlis, Malaysia.

Keywords:

Image Processing, MLP, P. Knowlesi Malaria, Recognition, Thin Blood Smears,

Abstract

Plasmodium (P.) Knowlesi is a fifth species of the malaria parasite in the world that caused a serious health problem. Current information suggests that P. Knowlesi is spread to people when an Anopheles mosquito infected by a monkey then bites and infects human (zoonotic transmission). Early identification of P. Knowlesi Malaria is an important step to an effective treatment. P. Knowlesi Malaria identification process is usually carried out with a 100x magnification of thin blood smear using microscope observation. However, this process is time-consuming and very tedious and strenuous for the human eyes. It also has a problem to differentiate between trophozoite, positive control (schizont and gametocyte) and negative control (white blood cell (WBC) and artefact). To overcome these situations, a computer-aided diagnosis system is developed to automatically identifying trophozoite stages of P. Knowlesi Malaria as early identification species, positive control and negative control. The processes involved starting from image acquisition, image processing and recognition. For image processing method, the most important part is the segmentation where the Otsu’s method is utilised to obtain the region of interest (ROI) of the infected cell. The features consist of the size of infected cell and size of the object. Finally, the recognition method using Multilayer Perceptron (MLP) is applied. The results show that the proposed automatic procedure is capable of recognising the trophozoite stage of P. Knowlesi Malaria with an accuracy of 98.70% for training and 97.67% for testing, using MLP trained by Lavernberg Marquardt (LM).

Downloads

Published

2018-05-30

How to Cite

Kanafiah, S., Mashor, M., Mohamed, Z., Teh, J., Abdul Shukor, S., Zakaria, N., Lim, H., & Ali, H. (2018). An Intelligent Recognition Procedure for Trophozoite Stages of Plasmodium Knowlesi Malaria. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 10(1-16), 31–35. Retrieved from https://jtec.utem.edu.my/jtec/article/view/4071