Klasifikasi Citra Tumor Otak MRI Menggunakan Convolutional Neural Network Model InceptionResNet-V2

As the center of the nervous system, the brain plays an important role in controlling various bodily functions. Brain tumors, occur due to abnormal cell growth in the brain that can interfere with these functions and require appropriate treatment. This study uses Convolutional Neural Network (CNN) with InceptionResNet-V2 architecture to classify brain tumor types in MRI images. This model combines the inception module and residual network connections to extract complex features from images. The purpose of this study is to improve the accuracy and efficiency of brain tumor type classification by data augmentation and Contrast Limited Adaptive Histogram Equalization (CLAHE) technique to enhance image contrast. The dataset used is taken from the Kaggle website with the title "Brain Tumor Classification (MRI)", including four classes: no tumor, glioma tumor, meningioma tumor, and pituitary tumor. The results show that the InceptionResNet-V2 architecture with data augmentation and CLAHE technique significantly improves the accuracy of brain tumor type classification, achieving the highest accuracy of 98%.