Plasmodium classification on red blood cells image using multiclass support vector machines

S. F.Y.O. Pradini, A. Bustamam, Z. Rustam

Research output: Contribution to journalConference articlepeer-review


Classification methods have been frequently used in various aspects, including bioinformatics. One of it's purpose of this classification is to determine phase level of a disease. This research will classify the phase of plasmodium falciparum parasite which causes malaria. The disease is spread by an infected female Anopheles mosquito, which contains Plasmodium. The result of this research could be used to determine Plasmodium parasite phase in infected people's red blood cells. The purpose of this research is to discover the success rate of Multiclass Support Vector Machines method and analyze it in order to predict the parasite phase levels. The data of this study is image data of red blood cells which were infected by three kinds of Plasmodium falciparum parasite levels. In the process, this study will be using Canopy as Integration Development Environments of phyton programming language. From 112 trials, the highest number of accuracy is 87.5% for Multiclass Support Vector Machines one vs rest method which used the 4-fold cross-validation with C=1 as parameter for linear kernel.

Original languageEnglish
Article number032020
JournalJournal of Physics: Conference Series
Issue number3
Publication statusPublished - 3 Jul 2020
Event6th International Conference on Mathematics, Science, and Education, ICMSE 2019 - Semarang, Central Java, Indonesia
Duration: 9 Oct 201910 Oct 2019


Dive into the research topics of 'Plasmodium classification on red blood cells image using multiclass support vector machines'. Together they form a unique fingerprint.

Cite this