Segmentasi Citra Nukleus Sel Kanker Serviks Menggunakan Otsu Thresholding Dan Morfologi Closing

  • Nugrohoputri R
  • Desiani A
  • Wahyudi Y
  • et al.
N/ACitations
Citations of this article
36Readers
Mendeley users who have this article in their library.

Abstract

Abstrak Kanker serviks dapat disembuhkan jika kelainan dan gejalanya dapat terdeteksi sedini mungkin. Pekerjaan diagnosa manual membutuhkan banyak waktu dan teknisi sitologi yang terampil. Sistem pengelolaan citra medis ini diharapkan dapat mengurangi faktor-faktor subjektivitas dan kesalahan diagnosa kanker serviks. Metode yang digunakan pada penelitian ini adalah Otsu Thresholding dan Morfologi Closing. Penelitian ini dilakukan dalam dua tahap, yaitu preprocessing dan dilanjutkan dengan segmentasi menggunakan metode yang diusulkan. Penelitian diukur menggunakan parameter yang terlibat untuk mengukur kinerja pendekatan yang diusulkan berdasarkan akurasi, sensitivitas, spesifisitas, F1-Score, dan IoU. Hasil penelitian dari 20 sampel gambar menunjukkan rata-rata metode yang diusulkan, yaitu Otsu Thresholding dan Morfologi Closing mampu melakukan perbaikan citra dan segmentasi citra nukleus sel kanker serviks dengan akurasi sebesar 0.8969, sensitivitas sebesar 0.8806, spesifisitas 0.9954, F1-Skor 0.9293, dan IoUq sebesar 0.8740. Penelitian ini menyimpulkan bahwa metode Otsu Thresholding dan Morfologi Closing mampu menghasilkan perbaikan citra dan segmentasi nukleus sel kanker serviks dengan lebih baik dibandingkan dengan metode-metode lainnya dari penelitian lain. Abstract Cervical cancer can be cured if the abnormalities and the symptoms can be detected as early as possible. Manual diagnostic requires a lot of time and skilled cytology technician. It is expected that this medical image management system can reduce the subjective factors and misdiagnoses of cervical cancer. This study uses Otsu Thresholding segmentation method and Morphology Closing. This research was performed by stages, first preprocessing and followed by segmentation using the proposed method. The study was measured using the parameters involved to measure the performance of the proposed approach based on accuracy, sensitivity, specificity, F1-Score, and IoU. The results of the 20 sample images showed that the average method proposed, namely Otsu Thresholding and Morphology Closing was able to improve the image and segmentation of cervical cancer cell nucleus images with accuracy of 0.8969, sensitivity of 0.8806, specificity of 0.9954, F1-Score of 0.9293, and IoU. of 0.8740. This study concludes that Otsu Thresholding and Closing Morphology methods are able to construct image improvements and segmentation of cervical cancer cell nuclei better than other methods from other studies.

Cite

CITATION STYLE

APA

Nugrohoputri, R. F., Desiani, A., Wahyudi, Y., Al-Filambany, M. G., Susanto, S., & Maiyanti, S. I. (2022). Segmentasi Citra Nukleus Sel Kanker Serviks Menggunakan Otsu Thresholding Dan Morfologi Closing. JSI: Jurnal Sistem Informasi (E-Journal), 14(1). https://doi.org/10.18495/jsi.v14i1.17098

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free