Pengaruh Preprocessing Terhadap Klasifikasi Diabetic Retinopathy dengan Pendekatan Transfer Learning Convolutional Neural Network

Abstract

Diabetic retinopathy (DR) is eye diseases caused by diabetic mellitus or sugar diseases. If DR is detected in early stage, the blindness that follow can be prevented. Ophthalmologist or eye clinician usually decide the stage of DR from retinal fundus images. Careful examination of retinal fundus images is time consuming task and require experienced clinicians or ophthalmologist but a computer which has been trained to recognize the DR stages can diagnose and give result in real-time manner. One approach of algorithm to train a computer to recognize an image is deep learning Convolutional Neural Network (CNN). CNN allows a computer to learn the features of an image, in our case is retinal fundus image, automatically. Preprocessing is usually done before a CNN model is trained. In this study, four preprocessing were carried out. Of the four preprocessing tested, preprocessing with CLAHE and unsharp masking on the green channel of the retinal fundus image give the best results with an accuracy of 79.79%, 82.97% precision, 74.64% recall, and 95.81% AUC. The CNN architecture used is Inception v3.

Analisis Sentimen Komentar Politik Di Media Sosial X Dengan Pendekataan Deep Learning

“…Convolutional Neural Network, pada prinsipnya, adalah suatu kategori Neural Network yang menggunakan operasi convolution sebagai pendekatan perkalian matriks, setidaknya terdapat pada satu lapisan tertentu. Struktur CNN melibatkan lapisan convolution dan pooling, memungkinkan proses ekstraksi fitur dari citra [13].…”

Comparison of CNN Models With Transfer Learning in the Classification of Insect Pests

“…The layers added to the end of the CNN are trained first, while the weights in the feature extraction section are kept unchanged. This allows the model to focus on training the new classification layer, enabling it to better learn the task at hand [15]. All pre-trained models used will have a Global Average Pooling layer, Dropout with a value of 0.3, and an output layer with 102 classes and a Softmax activation function added to them [4].…”

Pengaruh Fitur-fitur Instagram Terhadap Pemasaran Supp.Id pada Platform Social Media Instagram

“…Pemasaran online merupakan segala sesuatu yang dilakukan penjual untuk melakukan pemasaran produk atau jasa menggunakan fasilitas internet (Widyaya 2021).…”

“…Maka dapat disimpulkan dari hasil observasi bahwa memasarkan produk melalui fitur yang ada pada Instagram efektif, karena memudahkan pelanggan maupun owner. Berdasarkan hasil tersebut, maka adanya kesesuaian dengan teori pemasaran online yang dikemukakan oleh (Widyaya, 2021) mengemukakan bahwa pemasaran online memasarkan produk menggunakan fasilitas internet, serta mempergunakan media sosial untuk menghubungkan pelanggan dan owner.…”