Abstract :

Early And Accurate Diagnosis Of Breast Cancer Is Critical For Effective Treatment And Patient Survival. This Study Explores The Use Of Deep Learning Models For Breast Cancer Classification Using Pathological Images, Enhanced Through Data Augmentation Using CycleGAN—a Generative Adversarial Network Capable Of Translating Images Between Domains Without Paired Training Examples. The CycleGANbased Augmentation Method Helps In Overcoming The Limitations Of Small And Imbalanced Datasets, A Common Challenge In Medical Imaging. In This Research, Several Convolutional Neural Network Architectures, Including ResNet50, ResNet101, GoogleNet, VGG16, And Alex Net, Were Trained And Evaluated On Augmented Pathological Image Datasets. Key Performance Metrics Such As FScore, Recall, Precision, And Accuracy Were Calculated For Each Model To Determine Their Effectiveness In Breast Cancer Detection. Among The Tested Architectures, AlexNet Achieved The Highest Overall Performance With An Accuracy Of 88.76%, Followed By VGG16 And GoogleNet. The Results Demonstrate That CycleGAN-based Augmentation Significantly Improves Model Performance By Increasing The Diversity And Quantity Of Training Data, Enabling Better Generalization And Robustness In Classification. This Work Highlights The Potential Of Combining Generative Augmentation Techniques With Deep Learning For Enhanced Diagnostic Accuracy In Breast Cancer Pathology, Supporting The Development Of More Reliable Computer-aided Diagnosis Systems

Published:

26-7-2025

Issue:

Vol. 25 No. 7 (2025)

Page Nos:

829-833

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite