DETEKSI CARA PANDANG RUANG JANTUNG JANIN MENGGUNAKAN ARSITEKTUR YOU ONLY LOOK ONCE (YOLO)

KUSUMAWATI, SYERPRI WINDRIYA and Nurmaini, Siti (2022) DETEKSI CARA PANDANG RUANG JANTUNG JANIN MENGGUNAKAN ARSITEKTUR YOU ONLY LOOK ONCE (YOLO). Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011281823035.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011281823035_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
Preview	Text RAMA_56201_09011281823035_0002085908_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_56201_09011281823035_0002085908_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011281823035_0002085908_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011281823035_0002085908_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011281823035_0002085908_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (772kB) | Request a copy
	Text RAMA_56201_09011281823035_0002085908_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (900kB) | Request a copy
	Text RAMA_56201_09011281823035_0002085908_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

The most important thing to know about fetal development is the development of heart. The first step in automated analysis of fetal heart chambers is to locate the four important chambers of fetal heart and aorta on the ultrasound image. However, this is a very difficult task due to several important factors such as the large number of spots on the ultrasound image, the small size of the fetal heart chambers, and the small size of the fetal heart. The fetal heart chambers view of various types will be detected, which are Four Chamber View (FCV), Left Ventricular Outflow Tract (LVOT), Right Ventricular Outflow Tract (RVOT), and Three Vessel Trachea View (TVTV). This study uses the You Only Look Once (YOLO) V3 method which has very different approach using a classification or localizer to perform detection by applying the model to the image in several locations, scaling and assigning values to the image as material for detection. Initial data that has been divided into training and testing produces 8 models. The best model is model 8 based on the results of the largest mAP, which are 97.994% using Batch Size 3, Learning Rate 0.0001 and Epoch 100. The Unseen data test which was carried out using data outside of the data model with three trials, got the best performance among other results, that is the 3rd model with a total of 240 data having an mAP result of 25.733% using Batch Size 4, Learning Rate 0.001 and Epoch 100.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Citra Jantung Janin, Deteksi Ruang Jantung
Subjects:	T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Syerpri Windriya Kusumawati
Date Deposited:	16 Nov 2022 05:35
Last Modified:	16 Nov 2022 05:35
URI:	http://repository.unsri.ac.id/id/eprint/81890

Actions (login required)

View Item