Sistem Pengembangan Deteksi Kanker Prostat Berbasis Image Processing dengan Metode Convolutional Neural Network

Alimin (Prodi Sistem Dan Teknologi Informasi, Institut Teknologi Yadika Pasuruan, Indonesia)
Sigit Riyadi (Teknik Informatika, STMIK Yadika , Bangil, Indonesia)

Explore IT : Jurnal Keilmuan dan Aplikasi Teknik Informatika

In the medical field, the diagnosis of prostate cancer is done by biopsy and microscopy. This testing technique takes a long time for even a dermatologist and has its own risks during the biopsy process so that it greatly affects the results of diagnosis and healing. To overcome this problem, we need a method to detect prostate cancer automatically through microscopic images of prostate cells. In this study, the method used is Convolutional Neural Network (CNN) for the identification of prostate cancer through prostate cell images. After testing on 60 images, it was found that 4 images failed to identify. The results of this study indicate that the proposed method is able to identify prostate cancer using microscopic images of prostate cells with an accuracy rate of 93.3%.