Susatyo, Dityatama Alfatkhaan (2020) Pengurangan Speckle Noise Dan Peningkatan Kualitas Citra Dari Echocardiographic Untuk Pendeteksian Awal Pada Penyakit Jantung. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
07311540000016-Undergraduate_Thesis.pdf Download (2MB) | Preview |
Abstract
Penggunaan pencitraan ultrasonografi dalam diagnosis medis telah ditentukan dengan baik karena sifatnya yang non-invasive, biaya rendah, kemampuan membentuk pencitraan secara real time, dan kemajuan yang terus menerus berkembang dalam kualitas citra yang dihasilkan. Namun, dalam pengambilan citra ultrasonografi masih terdapat noise yang menyebabkan penurunan dari kualitas citra yang dihasilkan, salah satu yang paling mengganggu adalah speckle noise. Speckle noise telah lama diketahui sebagai suatu faktor yang membatasi pendeteksian dari citra ultrasound, terutama pada hasil citra dengan kontras yang rendah. Terdapat beberapa metode yang dapat digunakan dalam mereduksi noise yang dihasilkan, namun pada penelitian ini metode yang digunakan yaitu Hybrid Oriented Speckle Reducing Anisotropic Diffusion (HOSRAD) yang merupakan gabungan dari metode Relaxed Median Filter dan metode Oriented Speckle Reducing Anisotropic Diffusion (OSRAD). Serta apabila dokter ingin mendapatkan batasan-batasan dari tepian dan pergerakan katup citra jantung dari hasil filter, maka digunakanlah metode Canny Edge Detection. Terdapat beberapa parameter yang digunakan sebagai analisa statistik kuantitatif untuk melihat perbedaan dari citra ultrasonografi yang telah direstorasi dengan citra ultrasonografi asli yang dijadikan sebagai referensi tersebut. Beberapa parameternya yaitu Signal to Noise Ratio (SNR), Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Figure of Merit (FOM), dan Structural Similarity quality Index (SSIM). Setelah berbagai percobaan didapatkan hasil restorasi yang citra yang paling baik dengan nilai gamma sebesar 0.5 pada tahap gamma correction, nilai time step sebesar 0.1 pada tahap OSRAD, dan Batasan atas serta bawah sebesar 5 dan 9 pada tahap Relaxed Median Filter. Sehingga didapatkan hasil metode HOSRAD dengan nilai MSE dan PSNR sebesar 70.13 dan 29.67 dengan iterasi sebanyak 5 kali. Hasil metode yang diajukan terbukti dapat meningkatkan kualitas citra jantung yang dari semula memiliki MSE dan PSNR sebesar 76.123 dan 29.31.
====================================================================================================================
The use of ultrasound imaging in medical diagnosis is well established due to its non-invasiveness, low cost, capability of forming real time imaging, and continuing improved image quality. However, in taking ultrasound images there is still a noise that causes the decreases of the ultrasound image quality, one of the most disturbing one is speckle noise. Speckle noise has long been known as a factor that limits the detection of ultrasound images, especially in low contrast images. There are several methods that can be used in reducing noise produced by ultrasound images, in this paper the method used is Hybrid Oriented Speckle noise Reducing Anisotropic Diffusion (HOSRAD) which is a combination of the Relaxed Median Filter method and the Oriented Speckle Reducing Anisotropic Diffusion (OSRAD) method. And if the doctor wants to obtain the edges of heart boundaries and heart valves movement, the method used is Canny Edge Detection. There are several parameters used as quantitative statistical analysis in order to see the differences in ultrasound images that has been restored with the original ultrasound images used as a reference, namely Signal to Noise Ratio (SNR), Peak Signal to Noise Ratio (PSNR), and Mean Square Error (MSE), Figure of Merit (FOM), and Structural Similarity quality Index (SSIM). After various experiments, the results for the best image restoration with a gamma value of 0.5 at the gamma correction process, a time step of 0.1 at the OSRAD process, and upper and lower boundaries of 5 and 9 at the Relaxed Median Filter. So, we get the results of the HOSRAD method with MSE and PSNR values of 70.13 and 29.67 with 5 iterations. The results of the proposed method are proven to be able to improve the quality of heart images which from the beginning had MSE and PSNR of 76,123 and 29.31.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Anisotropic diffusion, Image enhancement, Pencitraan echocardiography, Speckle noise, Speckle reduction, Anisotropic diffusion, Echocardiography imaging, Image enhancement, Speckle noise, Speckle reduction |
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer programming. R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques |
| Divisions: | Faculty of Electrical Technology > Biomedical Engineering > 11410-(S1) Undergraduate Thesis |
| Depositing User: | Dityatama Alfatkhaan Susatyo |
| Date Deposited: | 27 Aug 2020 08:12 |
| Last Modified: | 03 Jun 2023 15:26 |
| URI: | http://repository.its.ac.id/id/eprint/79515 |
Actions (login required)
 |
View Item |