Breast cancer tissue stiffness assessment using transfer learning ultrasound elastography: A phantom study

Abstract

In this work, a phantom study is performed to investigate the feasibility of tissue stiffness assessment of breast cancer masses using transfer learning ultrasound elastography. A transfer learning ultrasound elastography model is developed to classify the breast masses into quantifiable Young’s modulus (kilopascals, kP) values regimes. The transfer learning model combines features of ultrasound elastography elastograms and images from Google’s deep learning models. The elastogram features used in the training and validation of the transfer learning model were obtained from a calibrated elastography phantom with inclusions having various Young’s moduli. The model was tested on a breast phantom with spherical inclusions. Test results show that the transfer learning model yields greater than 88% validation accuracy. Further results and limitations of the transfer learning techniques will be presented and discussed.