Deteksi Kanker Serviks Otomatis Berbasis Jaringan Saraf Tiruan LVQ dan DCT

Dhimas Arief Dharmawan

Abstract


Abstract—Cervical cancer has became the common women
disease in the world. Mostly, cervical cancer has been already
known lately, because it is very dificult to detect this in early
stage. In this work, a computer based software using Learning
Vector Quantization (LVQ) has been designed as the early
cervical cancer detection aid tool. There are six methods before
the detection is performed, namely preprocessing, contrast
stretching, median filtering, morphology operation, image
segmentation, and Discrete Cosine Transform based feature
extraction. In tihis work, 73 cervical cell images that consist of 50
normal images and 23 cancer images are used. 35 normal images
and 14 cancer images are used to train the LVQ. Then, 23
normal images and 9 cancer images are used in the testing
process. Our results show 88,89 % cancer image can be detected
correctly (sensitivity), 100 % normal image can be detected
corerctly (specificity), and 95,83 % for overall detection
(accuracy).
Intisari—Kanker serviks telah menjadi penyakit yang banyak
diderita kaum wanita di dunia. Secara umum, kanker serviks
baru terdeteksi setelah memasuki stadium lanjut, sebab kanker
ini sulit teramati pada stadium awal. Pada penelitian ini
dirancang perangkat lunak dengan jaringan saraf tiruan
Learning Vector Quantizatin (LVQ), sebagai alat bantu deteksi
kanker serviks. Sebelum dideteksi, dilakukan pengolahan citra
terhadap citra sel serviks, yaitu preprocessing, peregangan
kontras, median filter, operasi morfologi, segmentasi, dan
ekstraksi fitur dengan Discrete Cosine Transform (DCT). Citra
sel serviks yang digunakan berjumlah 73 buah yang terdiri atas
lima puluh buah citra sel normal dan 23 buah citra sel kanker.
Proses pelatihan LVQ menggunakan 35 buah citra sel normal
dan empat belas buah citra sel kanker. Proses pengujian LVQ
menggunakan 15 buah citra sel normal dan sembilan buah citra
sel kanker. Dari hasil pengujian, didapatkan nilai sensitivitas,
spesifisitas, dan akurasi sebesar 88,89 %, 100 %, dan 95,83 %.

Full Text:

PDF


DOI: http://dx.doi.org/10.22146/jnteti.v3i4.114

Refbacks

  • There are currently no refbacks.


Copyright (c) 2016 Jurnal Nasional Teknik Elektro dan Teknologi Informasi (JNTETI)

JNTETI (Jurnal Nasional Teknik Elektro dan Teknologi Informasi)

Departemen Teknik Elektro dan Teknologi Informasi, Fakultas Teknik Universitas Gadjah Mada
Jl. Grafika No 2. Kampus UGM Yogyakarta 55281
+62 274 552305
jnteti@ugm.ac.id