Exploring the Potential of ResNet50 and YOLOv8 in Improving Breast Cancer Diagnosis: A Deep Learning Perspective
Abstract
Breast cancer is the most common cancer among women worldwide, and early detection is crucial for improving survival rates. However, traditional manual diagnosis of breast cancer from histopathological images is time-consuming and subjective. In this research, we explore the performance of two deep learning models, ResNet50 and YOLOv8, for binary classification of breast cancer histopathology images. The models are trained and tested on the BreakHis dataset, which contains 7,909 images of benign and malignant breast tumors. To address the class imbalance issue in the dataset, data augmentation and oversampling techniques are employed to increase the diversity and number of benign samples. The performance of the models is evaluated based on metrics such as accuracy, precision, recall, F1-score, and false negative rate. The results show that YOLOv8 outperforms ResNet50 in terms of accuracy and false negative rate, achieving 97.8% and 1.2%, respectively. The study demonstrates the effectiveness and efficiency of YOLOv8 in breast cancer classification, as well as its potential for real-time applications in medical image analysis.