Assessment of transfer learning ultrasound elastography: A breast cancer phantom study

In this work, a phantom study is performed to investigate the feasibility of quantitative 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. The transfer learning model combines features of B-mode images and elastograms from Google’s deep learning model AlexNet. The B-mode images and elastograms from a calibrated phantom with elastic inclusions are used to train and validate the model. Thereafter, the model is used to quantify Young’s modulus of inclusions from an uncalibrated breast phantom. The accuracy of the transfer learning results with and without the inclusion of the B-mode is discussed.