MAULA, RIZKI (2025) Analisis Kinerja Feature Descriptor pada Klasifikasi Penyakit Ginjal Melalui Analisis Citra Computerized Tomography (CT) Scan Organ Ginjal. S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.
![[img]](https://eprints.untirta.ac.id/style/images/fileicons/text.png) |
Text (Fulltext)
Rizki Maula_3332210011_Fulltext.pdf Restricted to Registered users only Download (2MB) |
|
Text (Bab 1)
Rizki Maula_3332210011_01.pdf Restricted to Registered users only Download (1MB) |
|
Text (Bab 2)
Rizki Maula_3332210011_02.pdf Restricted to Registered users only Download (600kB) |
|
Text (Bab 3)
Rizki Maula_3332210011_03.pdf Restricted to Registered users only Download (554kB) |
|
Text (Bab 4)
Rizki Maula_3332210011_04.pdf Restricted to Registered users only Download (429kB) |
|
Text (Bab 5)
Rizki Maula_3332210011_05.pdf Restricted to Registered users only Download (280kB) |
|
Text (Daftar Referensi)
Rizki Maula_3332210011_Ref.pdf Restricted to Registered users only Download (285kB) |
|
Text (Lampiran)
Rizki Maula_3332210011_Lamp.pdf Restricted to Registered users only Download (637kB) |
|
Text (Cek Plagiarisme)
Rizki Maula_3332210011_CP.pdf Download (9MB) |
Abstract
Kidney, as one of the organs of the human body, can experience disorders that can be caused by several diseases, including kidney stones, kidney tumors, and kidney cysts. These diseases can be detected by reading computerized tomography (CT) scan images. Artificial intelligence has been developed to classify kidney diseases through CT scan images, but there are several problems, namely the quality of the classification model, noise in the image, and the complexity of the image. Therefore, this study analyzed the accuracy of the model and the effectiveness of using feature descriptors in the process of classifying kidney disease through CT scan images. In this study, it is known that the Support Vector Machine (SVM) model with the Histogram of Oriented Gradient (HOG) feature is the best combination with a weighted average f1-score of 1 and effectively helps the model overcome the problem of misclassification of the actual tumor class in experiments without descriptors. In addition, the use of feature descriptors is also effective in reducing model complexity by observing the model training time, which is reduced by 14%–99,7%.
| Item Type: | Thesis (S1) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
|||||||||
| Uncontrolled Keywords: | Ginjal, Feature Descriptor, Computerized Tomography Scan, Artificial Intelligence, Klasifikasi Citra | |||||||||
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering | |||||||||
| Divisions: | 03-Fakultas Teknik > 20201-Jurusan Teknik Elektro | |||||||||
| Depositing User: | Rizki Maula | |||||||||
| Date Deposited: | 19 Jun 2025 02:31 | |||||||||
| Last Modified: | 19 Jun 2025 02:31 | |||||||||
| URI: | http://eprints.untirta.ac.id/id/eprint/49711 |
Actions (login required)
 |
View Item |