METODE ENSEMBLE PADA ARSITEKTUR RESNET-50, MOBILENET, DAN EFFICIENT NET DALAM PENENTUAN PENYAKIT DIABETIC RETINOPATHY PADA RETINA

AL-FILAMBANY, MUHAMMAD GIBRAN and Desiani, Anita and Suprihatin, Bambang (2022) METODE ENSEMBLE PADA ARSITEKTUR RESNET-50, MOBILENET, DAN EFFICIENT NET DALAM PENENTUAN PENYAKIT DIABETIC RETINOPATHY PADA RETINA. Undergraduate thesis, Sriwijaya University.

	Text RAMA_44201_08011281722048.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_44201_08011281722048_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (12MB) | Request a copy
Preview	Text RAMA_44201_08011281722048_0011127702_0026017102_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_44201_08011281722048_0011127702_0026017102_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (550kB) | Request a copy
	Text RAMA_44201_08011281722048_0011127702_0026017102_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (302kB) | Request a copy
	Text RAMA_44201_08011281722048_0011127702_0026017102_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011281722048_0011127702_0026017102_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (154kB) | Request a copy
	Text RAMA_44201_08011281722048_0011127702_0026017102_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Attribution Share Alike. Download (181kB) | Request a copy

Abstract

Diabetic rethinopathy (DR) is a diabetic that affects the retinal of eye and can be identified through retinal images. Retinal image identification process can be done by applying deep learning-based methods, one of which is Convolutional Neural Network (CNN). CNN has an architecture that can perform the images classification process, namely ResNet-50, MobileNet, and EfficientNet. The weaknesses possessed by each architecture can be overcome through the ensemble learning method which can combine the performance results of each single classification methods. This research applies ensemble learning method to improve performance results of the ResNet-50, MobileNet, and EfficientNet architectures in determining retinal DR disease. APTOS and EyEPACS is used in this research. The method used is data collection, training, testing, and evaluation in each architecture and ensemble learning. The result of ensemble learning evaluation outperform the other architecture in term of accuracy, F1-Score and cohen kappa with value of 93,3%, 93,42%, and 0.866 respectiveley. The best result for specifity are achieved by Resnet with 99,78% and EfficientNet achieved the best result for sensitivity with 96,2%. Based on the result, it can be concluded that the ensemble learning method porposed perform very well.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Diabetic Retinopathy, Ensemble Learning, MobileNet, ResNet-50, EfficientNet.
Subjects:	Q Science > QA Mathematics > QA8.9-QA10.3 Computer science. Artificial intelligence. Computational complexity. Data structures (Computer scienc. Mathematical Logic and Formal Languages
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	Muhammad Gibran Al Filambany
Date Deposited:	30 Jun 2022 02:12
Last Modified:	30 Jun 2022 02:12
URI:	http://repository.unsri.ac.id/id/eprint/64964

Actions (login required)

View Item