Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Medical Image Registration in Pap Smear Images for Early Identification of Cervical Cancer

Rektorat UDINUS 2023 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2023
Goals
Record
214b0543-6f6e-4296-98bf-8edcc74c969f
Published
2026-10-30

Universitas Dian Nuswantoro conducted research on medical image registration in Pap smear images for the early identification of cervical cancer, with Merlina, Nita; Noersasongko, Edi; Andono, Pulung Nurtantio; Soeleman, M Arief; Riana, Dwiza; and Naàm, Jufriadif participating as researchers. The activity addressed the difficulty of automatic visual interpretation caused by the complexity of cell structures and the high degree of overlap, which can produce poor image contrast. The source also identifies lighting differences, dye concentrations, drying air, excess blood, mucus, bacteria, and inflammation as imaging variables that can make interpretation more difficult.

The research proposed an approach model that combined basic image processing techniques with deep learning to segment the nucleus in overlapping cell images from Pap smears of cervical cancer patients. The purpose was to improve the identification accuracy of Pap smear images using the RepomedUNM public data. The reported performance included a lowest RMSE value of 0.2024253, a lowest PSNR of 0.04009707, and a highest PSNR of 65.3826018 dB. The source states that the study can be used as a reference for identifying the cervical cancer nucleus as medical image registration for patients.

This UDINUS research contributes to the health research objective by addressing image-based identification of a cervical cancer-related cellular structure through image processing and deep learning.

Reference https://www.scopus.com/pages/publications/85161092641?origin=resultslist