Segmentasi Citra Imunohistokimia Kanker Payudara Untuk Menentukan Reseptor Estrogen Menggunakan Fitur Bentuk

Aisyi, Rohadatul (2022) Segmentasi Citra Imunohistokimia Kanker Payudara Untuk Menentukan Reseptor Estrogen Menggunakan Fitur Bentuk. Other thesis, Universitas Islam Riau.

[img] Text
173510560.pdf - Submitted Version
Restricted to Repository staff only

Download (2MB) | Request a copy

Abstract

Immunohistochemical examination can detect the types of hormone receptors, namely estrogen and progesterone receptors which are useful for determining the response of patients receiving hormonal therapy. The bonding reaction that occurs in the staining of breast cells using Hematoxylin (H) dye will produce a blue color indicating that the ER/PR is negative and the staining with Diaminobenzidine (DAB) will produce a brown color on the cells indicating that the ER/PR is positive. Color separation in IHC images using the color deconvolution method produces images based on cell color, namely blue on IHC H images and brown on positive DAB images. Marker-watershed segmentation by morphological processing is helpful in the separation of overlapping cells. The shape feature is useful for selecting the shape of cancer cells and separating non-nuclear cells which makes it easy to calculate the number of positive IHC H and DAB cells. Based on tests using the confusion matrix, the average sensitivity value is 76.43% and the average sensitivity value is 20.74%.

Item Type: Thesis (Other)
Contributors:
ContributionContributorsNIDN/NIDK
SponsorYulianti, AnaUNSPECIFIED
Subjects: Q Science > QA Mathematics > QA76 Computer software
T Technology > T Technology (General)
Divisions: > Teknik Informatika
Depositing User: Luthfi Pratama ST
Date Deposited: 10 Jan 2023 04:40
Last Modified: 10 Jan 2023 04:40
URI: http://repository.uir.ac.id/id/eprint/19353

Actions (login required)

View Item View Item