Aplikasi Model Kontur Aktif Berbasis Tepi pada Segmentasi Citra Single Photon Emission Computed Tomography Tiroid

Rahmandinof, Arif (2019) Aplikasi Model Kontur Aktif Berbasis Tepi pada Segmentasi Citra Single Photon Emission Computed Tomography Tiroid. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 01111540000090-Undergraduate Thesis.pdf Restricted to Repository staff only Download (897kB) | Request a copy

Abstract

Telah diselesaikan tugas akhir tentang Aplikasi Model Kontur Aktif Berbasis Tepi pada Segmentasi Citra Single Photon Emission Computed Tomography Tiroid dalam rangka mendapatkan segmentasi citra fisiologi tiroid. Percobaan dilakukan menggunakan data asli pasien dari instalasi kedokteran nuklir, yaitu citra SPECT tiroid hasil pemindaian kamera gamma dengan intensitas yang bervariasi, tergantung pada serapan Tc99m-Sestamibi yang terinjeksi. Data input yang digunakan dalam percobaan ini berupa citra grayscale, citra yang sudah diberi mark, dan citra ground truth tiroid pasien. Data tersebut diolah dengan menggunakan model kontur aktif berbasis tepi dari metode level-set. Fungsi edge-stop dan klasifikasi k-Nearest Neighbours digunakan untuk mengatasi citra dengan batas yang tidak jelas. Hasil penelitian menunjukkan bahwa metode yang diajukan dapat digunakan untuk memperoleh segmentasi citra yang diinginkan. Dalam rangka mendapatkan citra hasil segmentasi yang bagus, permasalahan inisialisasi awal diselesaikan dengan cara penggunaan foreground yang sedikit berbeda sehingga lebih mempermudah perluasan kontur. Semakin besar jumlah iterasi yang digunakan maka hasil segmentasi semakin mendekati batas yang diinginkan. Berdasarkan citra dengan obyek yang sudah tersegmentasi, dapat diperoleh ukuran luas tiroid yang menyerap radiofarmaka Tc99m-Sestamibi.
==================================================================================================================================
A final project on the Application of Edge-Based Active Contour Model for Thyroid Single Photon Emission Computed Tomography Image Segmentation has been completed in order to obtain the image segmentation of thyroid physiology. Experiments were conducted using real patient data from the nuclear medicine installation, specifically thyroid SPECT images obtained from gamma camera scanning with varying intensities, depending on the uptake of injected Tc99m-Sestamibi. The input data used in this experiment consisted of grayscale images, marked images, and patient thyroid ground truth images. The data were processed using an edge-based active contour model from the level-set method. An edge-stop function and k-Nearest Neighbors classification were utilized to handle images with unclear boundaries. The results showed that the proposed method can be used to achieve the desired image segmentation. In order to obtain good segmentation results, the initial initialization problem was resolved by using a slightly different foreground, thereby facilitating contour expansion. A higher number of iterations resulted in segmentation outcomes that closer approached the desired boundaries. Based on the images with segmented objects, the surface area measurement of the thyroid absorbing the Tc99m-Sestamibi radiopharmaceutical can be determined.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Edge-stop Function, Inisialisasi Awal, Kontur Aktif Berbasis Tepi, Segmentasi Citra, SPECT Tiroid
Subjects:	Q Science > QC Physics > QC20.7.F67 Fourier transformations Q Science > QC Physics > QC795.5 Radioactivity and radioactive Instruments and apparatus (General) Q Science > QH Biology R Medicine > R Medicine (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Faculty of Mathematics and Science > Physics > 45201-(S1) Undergraduate Thesis
Depositing User:	Arif Rahmandinof
Date Deposited:	20 Jul 2026 03:40
Last Modified:	20 Jul 2026 03:40
URI:	http://repository.its.ac.id/id/eprint/66888

Actions (login required)

View Item