Andhika Putra Setianto (2021) APLIKASI PENGUKURAN OTOMASTIS LUAS JANTUNG DARI GAMBAR CHEST X – RAY MENGGUNAKAN METODE U- NET DEEP LEARNING. S1 thesis, Universitas Muhammadiyah Yogyakarta.
Halaman Judul.pdf
Download (265kB)
Lembar Pengesahan.pdf
Restricted to Registered users only
Download (303kB)
Abstrak.pdf
Restricted to Registered users only
Download (138kB)
Bab I.pdf
Download (220kB)
Bab II.pdf
Restricted to Registered users only
Download (477kB)
Bab III.pdf
Restricted to Registered users only
Download (279kB)
Bab IV.pdf
Restricted to Registered users only
Download (292kB)
Bab V.pdf
Restricted to Registered users only
Download (35kB)
Daftar Pustaka.pdf
Restricted to Registered users only
Download (159kB)
Lampiran.pdf
Restricted to Registered users only
Download (420kB)
Naskah Publikasi.pdf
Restricted to Registered users only
Download (422kB)
Full Text.pdf
Restricted to Repository staff only
Download (1MB)
Abstract
Heart health is a fundamental human right and an essential element of global health justice. In this increasingly advanced era, every activity becomes very easy. This is due to the development of science and technology and information. But, so far some activities are still done manually. So it is necessary to have a computerized system design innovation. Calculation of the area of the heart in the human body can be done by combining image processing and deep learning techniques. Deep learning is a branch of science from machine learning. One part of the image processing process is image segmentation. This research uses the U-Net segmentation method to detect various stages of heart area calculation. U-Net can perform image segmentation process precisely and work with little training dataset. The population in this study were chest x-rays on the human body taken through the Kaggle website. The sample used is a human heart. The total images obtained are 800 chest x-ray images. The results of this study indicate that the level of accuracy in the training data obtained from the model with U-Net architecture is 0.98, while the results of the testing process are still doing calculations manually. The U-Net model in this study uses an input shape measuring 256x256, a kernel size of 3 x 3, and the number of epochs of 50
Item Type: | Thesis (S1) |
---|---|
Divisions: | Fakultas Teknik > Teknologi Informasi S1 |
Depositing User: | Unnamed user with email robi@umy.ac.id |
Date Deposited: | 14 Dec 2021 02:53 |
Last Modified: | 14 Dec 2021 02:53 |
URI: | https://etd.umy.ac.id/id/eprint/5736 |