Nirmalasari, Nur Indah (2019) Bayesian Spatially Constrained Fernandez-Steel Skew Normal Mixture Model untuk Segmentasi Citra MRI Tumor Otak. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211745000030-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Pemindaian otak dengan menggunakan Magnetic Resonance Imaging (MRI) merupakan salah satu cara untuk mendeteksi tumor otak. MRI lebih unggul dalam mendeteksi gangguan pada jaringan lunak dibandingkan alat radiologi lainnya. Namun pada citra MRI terdapat noise yang muncul secara random sehingga menyebabkan kesulitan dalam mendeteksi tumor secara tepat. Segmentasi citra dapat membantu mendiagnosa lokasi tumor otak dengan cara memisahkan Region of Interest (ROI) yaitu area tumor degan bagian lainnya. Gaussian Mixture Model (GMM) merupakan metode yang paling sering digunakan untuk segmentasi citra, namun distribusi Gaussian kurang mampu untuk menjelaskan pola data citra MRI. Selain itu, GMM tidak mempertimbangkan dependensi spasial antar pixel sehingga mengakibatkan kurangnya ketahanan terhadap noise. Penelitian ini dilakukan dengan menggunakan distribusi Fernandez Steel Skew Normal (FSSN) sebagai alternatif dari distribusi Gaussian pada GMM. Distribusi FSSN dapat mengakomodasi pola asimetris secara adaptif pada data citra MRI. Selanjutnya, untuk meningkatkan ketahanan terhadap noise dapat mempertimbangkan dependensi spasial antar pixel dengan menggunakan Markov Random Field (MRF) sebagai prior. Sehingga model yang digunakan dalam penelitian ini adalah Spatially Constrained FSSN mixture model (Sc-FSSNMM). Hasil segmentasi dengan menggunakan Sc-FSSNMM memberikan hasil segmentasi yang hampir sama dengan FSSNMM namun lebih baik dalam mengurangi noise yang berada jauh dari ROI.
=================================================================================================================================
Brain scanning using Magnetic Resonance Imaging (MRI) can be used to detect the brain tumor. MRI could detect the soft tissue abnormalities better than the other radiological devices. However, the noise in the image of the MRI sometimes appears randomly, so that it is difficult to detect the tumor more precisely. The image segmentation, therefore, is seeded to be able to diagnose the location of the brain tumor by separating the tumor as the Region of Interest (ROI) from others region. Gaussian Mixture Model (GMM) is commonly used for image segmentation. This method, however, frequently provides a poor result since it is less able to explain the skew pattern of MRI data. Moreover, the GMM is not considering the spatial dependencies between pixel, therefore it is less robust to noise. This study tries to employ the Fernandez Steel Skew Normal (FSSN) distribution as the replacement of the Gaussian in the GMM. The FSSN distribution could accommodate the symmetrical and even an asymmetrical pattern of the MRI data adaptively. In order to increase the noise robustness, the spatial dependencies with Markov Random Field (MRF) are used as a prior in Bayesian Markov chain Monte Carlo. The proposed model is called as the Spatially Constrained FSSN mixture model (Sc-FSSNMM). The results show that by applying the Sc-FSSNMM, the segmentation result is similiar with FSSNMM, but more robust to noise.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.542 Nir b-1 2019 |
| Uncontrolled Keywords: | Bayesian, Fernandez-Steel Skew Normal, Markov Random Field, Mixture Model, Segmentasi |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.7 Estimation T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Nur Indah Nirmalasari |
| Date Deposited: | 30 May 2024 03:29 |
| Last Modified: | 30 May 2024 03:29 |
| URI: | http://repository.its.ac.id/id/eprint/64459 |
Actions (login required)
 |
View Item |