| Penulis/Author |
DONI SETYAWAN (1); Prof. Drs. Retantyo Wardoyo, M.Sc., Ph.D. (2) ; Moh. Edi Wibowo, S.Kom.,M.Kom., Ph.D. (3); Prof. dr. E. Elsa Herdiana Murhandarwati, M.Kes., Ph.D (4) |
| Abstrak/Abstract |
Malaria is a disease caused by Plasmodium parasites transmitted through the bites of female anopheles-mosquito that infect the human red blood cells (RBC). The standard malaria diagnosis is based on manual examination of a thick and thin blood smear, which heavily depends on the microscopist experience. This study proposed a system that can identify the life stages of plasmodium falciparum, i.e., ring, trophozoite, schizont, and gametocyte in human RBC. The system consists of five parts, namely image acquisition, preprocessing, image segmentation, feature extraction, and classification. The image preprocessing process is done using contrast stretching continued with HSV conversion to get the saturation and value channel, and then median filtering to remove the noise. Image segmentation using global intensity and Otsu thresholding. The feature extraction process utilizes a combination of texture and morphological features, then using neural networks to classify the life stages of Plasmodium. The results show that the proposed method achieves an accuracy of 80.95 %, a sensitivity of 81 %, and a specificity of 88.89 %, thus improving decision-making for malaria diagnosis. |
