Breast cancer is a prevalent and life-threatening disease, and early detection and treatment are crucial for improving patient outcomes. Neoadjuvant therapy (NAT) is a common treatment approach for breast cancer, but predicting its effectiveness can be challenging. This study aims to develop a predictive model that integrates traditional radiomics and 3D deep learning features from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to accurately predict the response to NAT in breast cancer patients.
The study utilized a dataset of 234 breast cancer patients who underwent NAT. Traditional radiomics and 3D deep learning features were extracted from the early and peak phases of DCE-MRI images. These features were then integrated and selected using statistical methods to build predictive models. The models were evaluated using various performance metrics, and the results showed that the combined model integrating radiomics and deep learning features from both early and peak phases of DCE-MRI achieved the best accuracy and AUC values.
The findings suggest that integrating multi-phase imaging and diverse features can enhance the predictive accuracy of NAT response in breast cancer. The constructed model has the potential to guide personalized treatment strategies and improve patient care. Further research and validation with larger datasets are recommended to enhance the stability and generalizability of the model. This study provides valuable insights into the application of radiomics and deep learning in breast cancer treatment and opens up new possibilities for personalized medicine.
