DETEKSI DEFECT SEPTUM JANTUNG JANIN BERBASIS CITRA 2 DIMENSI MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS

SAPITRI, ADE IRIANI and Nurmaini, Siti and Sukemi, Sukemi (2020) DETEKSI DEFECT SEPTUM JANTUNG JANIN BERBASIS CITRA 2 DIMENSI MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS. Master thesis, Sriwijaya University.

	Text RAMA_55101_09042621923001.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (724kB) | Request a copy
	Text RAMA_55101_09042621923001_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (22MB) | Request a copy
Preview	Text RAMA_55101_09042621923001_0002085908_0003126664_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (724kB) | Preview
	Text RAMA_55101_09042621923001_0002085908_0003126664_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (988kB) | Request a copy
	Text RAMA_55101_09042621923001_0002085908_0003126664_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55101_09042621923001_0002085908_0003126664_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55101_09042621923001_0002085908_0003126664_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (117kB) | Request a copy
	Text RAMA_55101_09042621923001_0002085908_0003126664_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (176kB) | Request a copy
	Text RAMA_55101_09042621923001_0002085908_0003126664_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy

Abstract

Congenital heart disease is a common disease that can be life-threatening. CHD has an important role in knowing the early diagnosis of the heart, especially the fetus. Medical image analysis is one of the topics that can support the diagnosis process, especially the occurrence of septal defects. The image analysis process can be done by segmenting, detecting, and classifying it. This is the main key in carrying out the analysis process in diagnosing diseases of defects. Convolutional neural network (CNN) is a deep learning technique that is often used, especially in image analysis. RCNN mask is a CNN architecture that can perform segmentation, detection, and classification processes simultaneously called instance segmentation (Johnson, 2018). The proposed approach uses CNN with the RCNN Mask architecture using fetal cardiac septal defect data. The results showed that the model performance obtained was 97.46% mean Average Precision (mAP), 77.49% Intersection over Union (IoU), and 87.22% Dice Score Similarity (DSC).

Item Type:	Thesis (Master)
Uncontrolled Keywords:	Defect Septum, Ultrasound Image, Convolutional Neural Network,Mask RCNN.
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation. R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	09-Faculty of Computer Science > 55101-Informatics (S2)
Depositing User:	Users 10069 not found.
Date Deposited:	20 Jan 2021 06:51
Last Modified:	20 Jan 2021 06:51
URI:	http://repository.unsri.ac.id/id/eprint/40659

Actions (login required)

View Item