Breast Cancer Prediction Based on the CNN Models
Nuo Chen, Boyu Han, Zhixin Li, Haotian Wang · Highlights in Science Engineering and Technology · 2023
In modern society, the natural lifespan of an individual increased dramatically benefitting from advanced yet accurate methods of medical treatment. Though many diseases could be treated with a cure, the treatment of cancer has yet to be overcome. Related medical research has proven that the combination of accurate breast cancer diagnoses and treatments at an early stage could prevent the spread of cancer cells as it could increase a person's potential lifespan by a large margin. This research has conducted a comprehensive study on improving the efficiency of autonomous image recognition of breast cancer diagnosis using deep learning models. We use the most advanced CNN baseline models for image recognition, including VGG, ResNet, Efficient, etc. We also select two typical breast cancer datasets and tested the models on them to make our result more convincing. The final enhanced model of ResNet 101 can achieve a recognition rate of 89.98% for the benign and malignant samples.