Automated Image Analysis Method for p-vivax Malaria Parasite Detection in Thick Film Blood Images

Authors

  • Leonardo Yunda Universidad Santiago de Cali image/svg+xml
  • Andrés Alarcón Laboratory of Applied Remote Sensing and Image Processing - Universidad de Puerto Rico, Recinto Mayagüez.
  • Jorge Millán Sigma Biomedical, Hialeah, FL

DOI:

https://doi.org/10.18046/syt.v10i20.1151

Keywords:

Malaria, Thick film microscopy, Neuronal networks, principal component analysis.

Abstract

An image analysis method for Malaria parasite detection in thick film blood images is described. The developed method uses a combination of AGNES and Morphological Gradient techniques in the image segmentation stage. Wavelet-based feature extraction is followed by a neural network classification stage. Principal Component Analysis (PCA) is used to reduce the number of features and improve the performance of the neuronal network.  The true positive rate for determining a specific parasite was of 77.19%, while a 76.45% was obtained in determining at least a parasite in a microscopy image.

Author Biographies

  • Leonardo Yunda, Universidad Santiago de Cali

    Dean of Engineering School (Universidad Santiago de Cali, USC) and Director-Researcher in T@lebio, a Research Group within USC. Electronic Engineer (2000) and Master in Engineering (2006) from Universidad del Valle; Master in Telematics Engineering (2010) y Ph.D candidate in Telematics Engineering from Universidad de Vigo (España). Since 2001 until 2004 he worked as a researcher and development engineer in the Fraunhofer-IBMT Technology Center (Hialeah, FL), a branch of the Fraunhofer Institute for Biomedical Technology in St. Ingbert, Germany. His mayor areas of interest are Telemedicine and Digital Signal Processing.

  • Andrés Alarcón, Laboratory of Applied Remote Sensing and Image Processing - Universidad de Puerto Rico, Recinto Mayagüez.

    Electronic Engineer with a Master degree in Electronic Engineering (both from Universidad del Valle), presently linked to the Laboratory for Applied Remote Sensing and Image Processing (LARSIP), a multidisciplinary laboratory dedicated to the research and implementation of Remote Sensing, Hyperspectral Image Processing, Signal and Image Processing, Geographical Information Systems (GIS),Emergency Response Systems, Global Positioning Systems (GPS) technologies, Applied Electromagnetics and Bio-Optics, located within the Department of Electrical and Computer Engineering at the University of Puerto Rico, Mayagüez Campus. His mayor areas of interest are Communication’s Networks and Digital Signal Processing.

  • Jorge Millán, Sigma Biomedical, Hialeah, FL
    Physics (Universidad del Valle, 1988) with a Master degree in Physics (Universidad de Puerto Rico, 1993) and Electric Engineering (Pennsylvania University, 2000), and a Ph.D., in Biomedical Engineering (Universidad de Miami, 2006). Executive Director from Sigma Biomedical (Hialeah, FL), a contract research and development biomedical engineering company established to serve the scientific, engineering and regulatory needs of the biomedical sector, who provides custom solutions by a synergy of advanced instrumentation, biomedical signal processing, health telematics, medical imaging, medical device testing, software verification and validation, software and hardware development, mechanical engineering as well as medical device regulatory expertise.

Downloads

Published

2012-03-31

Issue

Vol. 10 No. 20 (2012)

Section

Original Research

License

This journal is licensed under the terms of the CC BY 4.0 licence (https://creativecommons.org/licenses/by/4.0/legalcode).