Breast Cancer Detection With Resnet50,Inception V3, And Xception Architecture

Abstract

This study proposes the implementation of three different deep learning models - ResNet50, InceptionV3, and Xception - for the purpose of identifying defects in breast thermographic images. The results of the study show that each architecture has its own unique strengths and weaknesses, and that the accuracy and performance of the models can be improved through the use of different optimizers and learning rates. This suggests that these models have the potential to be used in a variety of fields where high accuracy is required. Furthermore, the study compares the accuracy and performance of each model under different conditions. By varying the optimizers and learning rates used in each architecture, the study is able to determine which combination of parameters produces the highest accuracy and performance. Overall, this study suggests that the implementation of deep learning models is a promising approach for identifying defects in breast thermographic images, and that ResNet50, InceptionV3, and Xception are all viable options for achieving high accuracy and performance in this field. The study also emphasizes how crucial thorough parameter adjustment is for maximizing the precision and effectiveness of these models.