White Blood Cells play an important role as parts of the immune system by fighting against viruses, bacteria and potentially harmful foreign objects that enter the human body. The amount of white blood cells can indicate a certain disease or infection within the human body. This research aims to develop a system that can automatically detect and locate the location of white blood cells in a slide image that is stained or not stained. By not staining blood cell images, it can save time and resources that are normally used in white blood cell detection. This system is built using convolutional neural networks (CNN), a deep learning architecture. The CNN model is used for detecting white blood cells in stained images and is trained with 528 images and the model that is used for detecting white blood cells in unstained images is trained with 264 images. Bounding box regression is used to predict the location of white blood cells. The experiment test results show the detection accuracy for the stained images reach 53.85% and for the unstained images has 54.69% accuracy.
CITATION STYLE
Danny, D., Lina, L., & Chris, A. (2021). PENDETEKSIAN SEL DARAH PUTIH DARI CITRA PREPARAT DENGAN CONVOLUTIONAL NEURAL NETWORK. Jurnal Ilmu Komputer Dan Sistem Informasi, 9(1), 48. https://doi.org/10.24912/jiksi.v9i1.11557
Mendeley helps you to discover research relevant for your work.