Comparison of normalized confusion matrices for multiclass DR classification on the APTOS and Messidor-2 datasets using different preprocessing methods: Our pipeline with No Preprocessing (left), CLAHE preprocessing (middle), and Ben-Graham preprocessing (right).

Comparison of normalized confusion matrices for multiclass DR classification on the APTOS and Messidor-2 datasets using different preprocessing methods: Our pipeline with No Preprocessing (left), CLAHE preprocessing (middle), and Ben-Graham preprocessing (right).

<p><b>MobileViT</b> model on <i>APTOS2019</i> dataset shows excellent performance, with minimal misclassification and high precision-recall (AP=1.00). <b>DenseNet169</b> on the <i>Messidor-2</i> dataset achieves high accuracy.</p>

Saved in:

Bibliographic Details
Main Author:	Zarin Akter (22794652) (author)
Other Authors:	Jawad Ibn Ahad (22794655) (author), Md. Mutasim Farhan (22794658) (author), Riasat Khan (20135822) (author)
Published:	2025
Subjects:	Biotechnology Cancer Space Science Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified significantly impacts working refined embedding representation model &# 8217 gcn enhances embeddings enhanced deep learning different input regions convolutional neural network achieving superior performance main feature extractor extracted feature vectors diabetic retinopathy diagnosis quality assessment techniques graph convolutional networks facilitate clinical validation mobilevit obtained 96 90 %, f1 169 achieved 97 feature extraction using mobilevit external validation early diagnosis clinical environments 87 %, 169 proved vision impairment strengthen interpretability necessitating accurate leading cause eyepacs datasets eyepacs dataset extensive preprocessing entropy loss effective treatment class populations cam heatmaps best backbone art methods 63 %.
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

Cannot write session to /tmp/vufind_sessions/sess_l7u9c22r5ahqib4vtbrukgan39