DSpace Arşivi :: by Yazar "Iseri, Ismail" değerine göre listeleniyor

Listeleniyor 1 - 1 / 1

Thyroid Cancer Detection Using Convolutional Neural Networks
(Institute of Electrical and Electronics Engineers Inc., 2023) Ilkilic Aytac, Zeynep; Iseri, Ismail; Dandil, Besir
Thyroid cancer is one of the most common cancers among all cancer types, and early diagnosis is very important to prevent the disease from progressing. This study shows that using Convolutional Neural Network (CNN), a deep learning model for the diagnosis of thyroid cancer, can provide faster and more accurate results for the diagnosis of thyroid cancer, especially in clinical trials. At the same time, this study highlights the potential applications of CNN models in medical imaging and cancer diagnosis. The dataset used in the CNN model is composed of thyroid biopsy images and divided into two classes: benign and malignant. The use of the CNN model on high-resolution thyroid biopsy images demonstrated that the model can effectively classify and a test accuracy of 84% was achieved. The model provided high sensitivity and specificity compared to similar studies. The findings show that the diagnosis process of thyroid cancer can be accelerated by using the CNN model, thus providing a better treatment process for cancer patients. © 2023 IEEE.