KLASIFIKASI DAN VISUALISASI ENAM KELAS ABNORMALITAS JANTUNG JANIN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DAN SALIENCY MAP

AFRIANSYA, ROBI and Nurmaini, Siti (2023) KLASIFIKASI DAN VISUALISASI ENAM KELAS ABNORMALITAS JANTUNG JANIN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DAN SALIENCY MAP. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011181924010.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_56201_09011181924010_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (12MB) | Request a copy
	Text RAMA_56201_09011181924010_0002085908_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_56201_09011181924010_0002085908_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (930kB) | Request a copy
	Text RAMA_56201_09011181924010_0002085908_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (721kB) | Request a copy
	Text RAMA_56201_09011181924010_0002085908_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_56201_09011181924010_0002085908_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (126kB) | Request a copy
	Text RAMA_56201_09011181924010_0002085908_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (327kB) | Request a copy
	Text RAMA_56201_09011181924010_0002085908_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (617kB) | Request a copy

Abstract

This study presents and analyzes deep learning techniques to classify abnormalities in fetal heart images. This research compares four convolutional neural network (CNN) architectures to choose the best architecture with satisfactory results, and performs visualization using the Saliency Map method to provide insight regarding the part of the image that plays a role in the classification process. DenseNet121 architecture has the best classification performance with accuracy, sensitivity and specifications on validation data were 100%, 100%, and 100%, respectively and MobileNetV2 has the best classification performance with accuracy, sensitivity and specifications with score 90.2%, 65.7%, and 94.2% on unseen data, respectively. The proposed model yields satisfactory results, which means this model can support fetal cardiologists to interpret decisions to improve diagnostic abnormalities on fetal heart images.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Convolutional Neural Networks (CNN), Saliency Map, klasifikasi, Citra Jantung Janing, Abnormalitas.
Subjects:	T Technology > T Technology (General) > T57.6-57.97 Operations research. Systems analysis > T57.6.A2-Z General works Simulation Cf. QA76.9.C65 Computer science Cf. TA343 Engineering mathematics
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Robi Afriansya
Date Deposited:	07 Aug 2023 03:17
Last Modified:	07 Aug 2023 03:17
URI:	http://repository.unsri.ac.id/id/eprint/126021

Actions (login required)

View Item