A team of researchers from the University of Waterloo has developed a new method to predict the pathologic response of breast cancer patients to neoadjuvant chemotherapy. The researchers leveraged volumetric deep features from synthetic correlated diffusion imaging (CDIs), a type of magnetic resonance imaging (MRI), to predict the pathologic complete response (pCR) using a volumetric convolutional neural network.
The team found that the proposed method, called Cancer-Net BCa, provides enhanced pCR prediction performance compared to the current gold-standard imaging modalities. This breakthrough has the potential to help oncologists make more informed treatment recommendations for breast cancer patients and potentially improve patient outcomes. The method can also be extended to other applications of CDIs in the cancer domain to improve prediction performance further.
The tool uses a volumetric convolutional neural network to analyze MRI images obtained from a newly introduced imaging modality called synthetic correlated diffusion imaging (CDI). The study found that Cancer-Net BCa provided enhanced pCR prediction performance compared to traditional gold-standard imaging modalities and has the potential to aid oncologists in improving treatment recommendations for patients with breast cancer.
Cancer-Net BCa is a machine-learning model designed to help diagnose and treat breast cancer. In layperson’s terms, it is a computer program that uses data and patterns to identify the presence of breast cancer and predict the most effective treatments. Cancer-Net BCa aims to provide faster, more accurate, and more personalized diagnosis and treatment options for breast cancer patients.
This development demonstrates the potential of artificial intelligence in the medical field and its ability to aid medical professionals in making more informed decisions. Using advanced imaging techniques and machine learning algorithms, researchers are working towards creating a more precise and efficient method of diagnosing and treating cancer.