Segmentasi Jaringan Pembuluh Darah Pada MRI Menggunakan U-NET

Arjuna, Dzaky Hanif (2024) Segmentasi Jaringan Pembuluh Darah Pada MRI Menggunakan U-NET. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5024201078-Undergraduate_Thesis.pdf] Text
5024201078-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (7MB) | Request a copy

Abstract

Penelitian ini berfokus pada segmentasi jaringan pembuluh darah pada citra MRI menggunakan arsitektur U-Net. Segmentasi jaringan pembuluh darah merupakan bagian penting dari analisis citra medis dan berperan penting dalam diagnosis dan evaluasi penyakit neurologis. Penelitian ini bertujuan untuk meningkatkan akurasi dan efisiensi proses segmentasi dengan menerapkan model U-Net. U-Net, jaringan saraf tiruan konvolusional, dirancang untuk mengenali pola kompleks dalam struktur pembuluh darah. Dengan pelatihan menggunakan kumpulan data yang komprehensif, model ini dapat membedakan secara akurat antara pembuluh darah dan jaringan otak lainnya, sehingga menawarkan potensi besar untuk meningkatkan hasil diagnostik. Hasil awal menunjukkan bahwa model U-Net dapat menghasilkan segmentasi yang akurat, yang menjanjikan peningkatan signifikan dalam pemrosesan citra medis. Studi ini memberikan kontribusi penting terhadap pengembangan alat diagnostik berbasis artificial intelligence yang bertujuan untuk meningkatkan diagnosis dan pengobatan pasien dengan penyakit neurologis.
==============================================================================================================================
This research focuses on segmenting blood vessel networks in MRI images using U-Net architecture. Vascular tissue segmentation is an important part of medical image analysis and plays an important role in the diagnosis and evaluation of neurological diseases. This research aims to improve the accuracy and efficiency of the segmentation process by applying the U-Net model. U-Net, a convolutional artificial neural network, is designed to recognize complex patterns in blood vessel structures. By training on a comprehensive dataset, the model can accurately differentiate between blood vessels and other brain tissue, offering great potential to improve diagnostic yield. Preliminary results show that the U-Net model can produce accurate segmentation, which promises significant improvements in medical image processing. This study makes an important contribution to the development of artificial intelligence-based diagnostic tools aimed at improving the diagnosis and treatment of patients with neurological diseases.

Item Type: Thesis (Other)
Uncontrolled Keywords: Blood Vessel, Deep Learning, MRI Imaging, Segmentation, U-NET Citra MRI, Deep Learning, Pembuluh Darah, Segmentasi, U-Net
Subjects: R Medicine > R Medicine (General) > R857.M3 Biomedical materials. Biomedical materials--Testing.
R Medicine > R Medicine (General) > R858 Deep Learning
T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User: Dzaky Hanif Arjuna
Date Deposited: 24 Jul 2024 04:33
Last Modified: 24 Jul 2024 04:33
URI: http://repository.its.ac.id/id/eprint/108651

Actions (login required)

View Item View Item