Researchers found that a deep convolutional neural network, incorporating B-mode ultrasound and shear wave elastography data, had nearly an 88 percent sensitivity rate for predicting neoadjuvant chemotherapy response in patients with breast cancer.