Arifansyah, M. Olif (2025) Implementasi Algoritma Cnn Dalam Mendeteksi Penyakit Tumor Otak. Other thesis, Universitas Islam Riau.

Text 193510335.pdf - Submitted Version Restricted to Registered users only Download (4MB) | Request a copy

Abstract

Brain tumors are abnormal cell growths that pose life-threatening risks. One of the primary challenges in diagnosing brain tumors is the limited time and precision involved in analyzing medical imaging like MRI. To facilitate early detection, effective and accurate methods are essential to assist healthcare professionals in identifying tumors promptly. This research aims to address this issue by developing a brain tumor detection system based on Convolutional Neural Network (CNN) algorithms. Three CNN models— VGG16, Xception, and NASNet Mobile—were employed to classify brain tumor images. The dataset used consists of 4,000 images, comprising 2,000 tumor and 2,000 non-tumor images sourced from a public repository. The research process involved data preprocessing, model training, and performance evaluation using metrics such as accuracy, precision, recall, and F1-score. Experimental results revealed that the VGG16 model achieved the highest accuracy of 98%, followed by NASNet Mobile with 96.75%, and Xception with 96.75%. This study demonstrates that CNN algorithms can effectively classify brain tumor images, offering potential improvements in the speed and accuracy of medical diagnoses.

Actions (login required)

View Item