Deteksi Dini Retinopati Diabetik dengan Pengolahan Citra Berbasis Morfologi Matematika
Abstract: Diabetic retinopathy
is a complication caused by diabetes mellitus. Diabetic retinopathy, if not
handled properly can lead to blindness. A necessary step to prevent blindness
is early detection. Early detection can be done by finding the initial symptoms
that microaneurysm. In this research, a system made to detect diabetic
retinopathy using algorithms detection microaneurysm with mathematical
morphology. The algorithm is divided into three stages of preprocessing,
detecting candidate microaneurysm and postprocessing. In this research, the system will be made by
using a raspberry pi as the media. To see how well the system detects diabetic
retinopathy, the test will be done. in the tests performed, system obtained an
accuracy of 90%, sensitivity 90, and specificity of 55% using data diaretdb1.
While testing using data from e-ophtha obtained results with an accuracy of
70.5%, a sensitivity of 80% and a specificity of 60%.
Penulis: Lukman Heryawan
Kode Jurnal: jptinformatikadd170002