Framework Untuk Rekonstruksi Citra Pembuluh Darah Menggunakan Sistem Freehand Ultrasound 3D

Sunarya, I Made Gede (2020) Framework Untuk Rekonstruksi Citra Pembuluh Darah Menggunakan Sistem Freehand Ultrasound 3D. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
07111660010009-Disertation.pdf - Accepted Version
Restricted to Repository staff only

Download (14MB) | Request a copy

Abstract

Detection of vascular areas (blood vessels) using B-mode ultrasound images is required for automated applications such as registration and navigation in medical operations. Narrowing in the carotid artery increases the chance of su_ering a stroke in the future. The modalities that can be used to determine the location of the carotid artery are magnetic resonance, compute tomography and ultrasonography. The purpose of this study is to develop a framework for reconstructing blood vessel images using the 3D freehand ultrasound system in determining the exact location of the carotid artery blood vessels The 3D freehand reconstruction system of the ultraosund B-mode blood vessel lumen is a system that has input in the form of 2D B-mode ultrasound images of blood vessels and produces output in the form of 3D objects of blood vessels. The 3D freehand reconstruction of the ultrasound B-mode lumen vascular ultrasound image consists of three main stages: the extraction of the B-mode vascular ultrasound image feature, calibration of the ultrasound probe using N-_duncial phantom wire, 3D freehand reconstruction of the vascular lumen. RoI detection of carotid artery B-mode ultrasound image shows that SSD has an accuracy of 95 % and it is higher than the Hough Transform, Ellipse and Faster RCNN methods. The application of image enhancement with Gaussian _lters, Histogram Equalization and Median _lters gives an inuence in improving of detection accuracy. The proposed method processing time is _ 2 seconds. The 3D visualization process of blood vessel reconstruction consists steps of data acquisition, image enhancement, segmentation, outlier selection, elliptical feature _tting and visualization. The proposed procedure using image quality improvement in 3D visualization produces the highest accuracy compared to other comparison methods with an accuracy value of 99.41% and has the smallest standard deviation of 1.05. The best _tting polynomial is the order polynomial 22nd with an average error of 0.26303.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: ultrasound images, blood vessels, ultrasound image segmentation, 3D freehand reconstruction, 3D visualization citra ultrasound, pembuluh darah, segmentasi citra ultrasound, rekonstruksi freehand 3D, visualisasi 3D.
Subjects: Q Science > QA Mathematics > QA76 Computer software
Q Science > QA Mathematics > QA76.6 Computer programming.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20001-(S3) PhD Thesis
Depositing User: I Made Gede Sunarya
Date Deposited: 28 Aug 2020 06:14
Last Modified: 28 Aug 2020 06:14
URI: http://repository.its.ac.id/id/eprint/81501

Actions (login required)

View Item View Item