Solichah, Siti Azizah Nurul (2018) Segmentasi Citra Otak Magnetic Resonance Imaging (MRI) Menggunakan Gaussian Mixture Model (GMM) dengan Pendekatan Expectation Maximization (EM) untuk Membentuk Citra 3 Dimensi. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211645000008-Undergraduate_Theses.pdf - Accepted Version Download (7MB) | Preview |
Abstract
Tumor otak merupakan massa dari sel yang tidak diperlukan yang tumbuh di dalam saluran syaraf pusat. Tumor otak memberikan permasalahan klinis yang agak berbeda dengan tumor lain karena efek yang ditimbulkannya dan keterbatasan terapi yang dapat dilakukan. Salah satu peralatan medis yang digunakan untuk mendeteksi tumor otak yaitu Magnetic Resonance Imaging atau MRI. Untuk meningkatkan kualitas dari citra MRI tumor otak diperlukan suatu pendekatan segmentasi citra. Segmentasi objek di dalam citra bertujuan memisahkan region objek dengan wilayah latar belakang sehingga didapatkan hasil yang lebih mudah untuk dilakukan analisis selanjutnya. Banyak metode yang dikembangkan untuk segmentasi citra dalam ilmu statistika komputasi, salah satunya Model Based Clustering. Gaussian mixture model (GMM) merupakan model mixture yang paling sering digunakan dalam model based clustering khusunya pada segmentasi citra. GMM sering digunakan dalam klasifikasi hasil magnetic imaging resonance (MRI), positron emission tomography (PET), computed tomography (CT), dan hasil scan dalam dunia kedokteran lainnya. Kluster yang dapat memisahkan area tumor dengan area otak adalah 6 kluster sehingga model 3 dimensi dapat dibentuk untuk memperjelas penggambaran otak.
===================================================================================================
Brain tumor is such an unnecessary mass that growing inside central spine canal. It gives different clinical problem compared with other tumor cases because of its effects and limited therapy that applied. Primary brain tumor generally diagnosed in children and older adults. RSUD Dr. Soetomo Surabaya recorded the increasing of brain tumor cases and it occurs most frequently in women. One of imaging methods for diagnosis is Magnetic Resonance Imaging (MRI) that based on computer simulation of human body with approach of tomography. Imaging technique of MRI nearly complex because the image result based on the amount and optimal selecting of parameters. Therefore, image segmentation to enhance imaging result quality of MRI is needed. Object segmentation in an image is to separate object region and background region for easier further analytic. Gaussian mixture model frequently used for model based clustering especially image segmentation. GMM commonly used in image result clustering for Magnetic Resonance Imaging (MRI), positron emission tomography (PET), computed tomography (CT), and any others clinical scan. 6 clusters can divide ROI and Non-ROI so modeling on 3 dimension could be easier. 3 dimension modeling did on each slice (axial, coronal, sagittal).
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | tumor otak, Expectation Maximization, Gausssian mixture model, segmentasi citra, MRI. |
| Subjects: | R Medicine > RD Surgery |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Azizah N. S. Siti |
| Date Deposited: | 21 Jul 2021 23:33 |
| Last Modified: | 21 Jul 2021 23:33 |
| URI: | http://repository.its.ac.id/id/eprint/57757 |
Actions (login required)
 |
View Item |