Detection of Diabetic Retinopathy using Deep Learning and Transfer Learning Techniques with Oversampling to Address Imbalanced Dataset

Ranđelović, Teodora

dc.contributor.advisor	Rumetshofer, Elisabeth
dc.contributor.author	Ranđelović, Teodora
dc.date.accessioned	2026-01-06T11:54:52Z
dc.date.available	2026-01-06T11:54:52Z
dc.date.issued	2023
dc.date.submitted	2023-05-13
dc.identifier.uri	https://dspace.jcu.cz/handle/20.500.14390/48607
dc.description.abstract	The study aims to develop a system for detecting diabetic retinopathy using deep learning. In this study I have explored transfer learning with four distinct models and addressed the issue of an unbalanced dataset with oversampling. The final experiment achieved a significant improvement in accuracy and quadratic kappa score. The study highlights the potential of deep learning and the importance of addressing dataset imbalances for accurate results.	cze
dc.format	36p.
dc.format	36p.
dc.language.iso	eng
dc.publisher	Jihočeská univerzita	cze
dc.rights	Bez omezení
dc.subject	Diabetic Retinopathy	cze
dc.subject	Deep learning	cze
dc.subject	Transfer learning	cze
dc.subject	Convolutional neural network	cze
dc.subject	Image classification	cze
dc.subject	medical imaging	cze
dc.subject	diabetic macular edema	cze
dc.subject	retinal fundus photographs	cze
dc.subject	comparative analysis	cze
dc.subject	oversampling	cze
dc.subject	accuracy	cze
dc.subject	quadratic kappa score	cze
dc.subject	Diabetic Retinopathy	eng
dc.subject	Deep learning	eng
dc.subject	Transfer learning	eng
dc.subject	Convolutional neural network	eng
dc.subject	Image classification	eng
dc.subject	medical imaging	eng
dc.subject	diabetic macular edema	eng
dc.subject	retinal fundus photographs	eng
dc.subject	comparative analysis	eng
dc.subject	oversampling	eng
dc.subject	accuracy	eng
dc.subject	quadratic kappa score	eng
dc.title	Detection of Diabetic Retinopathy using Deep Learning and Transfer Learning Techniques with Oversampling to Address Imbalanced Dataset	cze
dc.title.alternative	Detection of Diabetic Retinopathy using Deep Learning and Transfer Learning Techniques with Oversampling to Address Imbalanced Dataset	eng
dc.type	bakalářská práce	cze
dc.identifier.stag	72690
dc.description.abstract-translated	The study aims to develop a system for detecting diabetic retinopathy using deep learning. In this study I have explored transfer learning with four distinct models and addressed the issue of an unbalanced dataset with oversampling. The final experiment achieved a significant improvement in accuracy and quadratic kappa score. The study highlights the potential of deep learning and the importance of addressing dataset imbalances for accurate results.	eng
dc.date.accepted	2023-06-02
dc.description.department	Přírodovědecká fakulta	cze
dc.thesis.degree-discipline	Bioinformatics	cze
dc.thesis.degree-grantor	Jihočeská univerzita. Přírodovědecká fakulta	cze
dc.thesis.degree-name	Bc.
dc.thesis.degree-program	Applied Informatics	cze
dc.description.grade	Dokončená práce s úspěšnou obhajobou	cze
dc.contributor.referee	Hofmarcher, Markus
dc.description.defence	<p>Committee: Konvička, Regl, Vohnout, Vohnoutova</p> <p>The student has presented her thesis within the time given. </p> <p>Questions:</p> <p>- Why did you not communicated well the experiment designs with your supervisor?</p> <p>- What if the retina is damaged from different diseases (not only diabetes) ? Are you able to detect it?</p> <p>- Your score is about 80%. What if you compare this to the accuracy of the doctor?</p> <p>- What is balanced accuracy in your work?</p>	cze