Skripsi

KOMBINASI ARSITEKTUR DENSENET BOTTLENECK LAYER DAN GATE RECURRENT UNITS (GRU) PADA KLASIFIKASI PENYAKIT GLAUKOMA MENGGUNAKAN CITRA RETINA

---

Glaucoma is a chronic eye disease that can lead to blindness. Glaucoma detection can be done by classification using the DenseNet architecture. DenseNet provides good model performance, but often suffers from overfitting. Bottleneck layers can be used to prevent overfitting by reducing the feature dimensions before entering deeper layers. However, reducing the feature dimension may lead to the loss of useful features. Another method that can reach features that have been skipped is the Gate Recurrent Units (GRU) architecture. GRU can update features in the input data by involving information about the overall state of the network. This research applies DenseNet architecture combined with bottleneck layer and GRU architecture. The results of research with retinal image datasets consisting of 3 classes obtained an accuracy value of 98.181%, sensitivity 97.32%, specificity 98.65%, f1-score 97.25%, and cohen's cappa 95.88%. The training graph of the method used proves that this study is able to overcome overfitting. Based on these results, it shows that the combination of DenseNet Bottleneck layer architecture and GRU is able to perform the classification task very well. Keyword : Glaucoma, Classification, retinal image, DenseNet, Bottleneck layers, GRU

---

Availability

Detail Information

Series Title

-

Call Number

T1566662024

Publisher

Indralaya : Prodi Ilmu Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam Universitas Sriwijaya., 2024

Collation

xii, 74 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.407

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Ilmu Matematika
Deep learning — Klasifikasi citra

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment