期刊名称：International Journal of Advances in Soft Computing and Its Applications
印刷版ISSN：2074-8523
出版年度：2020
卷号：12
期号：1
页码：35-48
出版社：International Center for Scientific Research and Studies
摘要：Malaria is caused by Plasmodium and transmitted by female Anophelesmosquitoes bite. Early detection of malaria has been performed usingmicroscopic thin blood smear images. Even so, in 2014, as many as 321 milliontimes microscopic examination, there were frequent misdiagnosis caused byhuman factors. Region of Interest (RoI) in image processing is part of theimage that has the highest information. The precise determination of the RoIarea can make the computer-based identification process work more efficiently,contribute better to the system, and also eliminate objects that are perceived tointerfere with the whole process. This study aims to detect parasites bydetermining the RoI area of two species of Plasmodium Ovale and PlasmodiumMalariae. The working principle of detection is using adaptive thresholding,colour segmentation between green channel (G) with hue channel (H) fromHSV colour space, and some morphological operation techniques. The dataused are digital microscopic images of thin blood smear. This research achievesthe sensitivity level of 91.6% and positive predictive value (PPV) of 89.1%. Itshows that performance of the proposed parasite detection method is reliable toassist doctor and contributes for developing the computer aided detection inPlasmodium cases.
关键词：HSV; Parasite Detection System; Plasmodium Malariae;Plasmodium Ovale; Region of Interest.
