A Data-Efficient Deep Learning Strategy for Tissue Characterization via Quantitative Ultrasound: Zone Training

Abstract

Deep learning (DL) powered biomedical ultrasound imaging is an emerging research field where researchers adapt the image analysis capabilities of DL algorithms to biomedical ultrasound imaging settings. A major roadblock to wider adoption of DL powered biomedical ultrasound imaging is that acquisition of large and diverse data-sets is expensive in clinical settings, which is a requirement for successful DL implementation. Hence, there is a constant need for developing data-efficient DL techniques to turn DL powered biomedical ultrasound imaging into reality. In this work, we develop a data-efficient deep learning training strategy for classifying tissues based on the ultrasonic backscattered RF data, i.e., quantitative ultrasound (QUS), which we named Zone Training. In Zone Training, we propose to divide the complete field of view of an ultrasound image into multiple zones associated with different regions of a diffraction pattern and then, train separate DL networks for each zone. The main advantage of Zone Training is that it requires less training data to achieve high accuracy. In this work, three different tissue-mimicking phantoms were classified by a DL network. The results demonstrated that Zone Training can require a factor of 2-3 less training data in low data regime to achieve similar classification accuracies compared to a conventional training strategy.

Calibrating Data Mismatches in Deep Learning-Based Quantitative Ultrasound Using Setting Transfer Functions

“…After acquiring ultrasound frames by either stable acquisition or free-hand acquisition, we extracted square data patches from the image frames whose sizes were 200 × 26 samples that correspond to square image patches whose sizes were 4 × 4 mm in physical dimensions, to be used in training, validation, and testing sets. The motivation behind patch extraction was described in our previous work through a clinical scenario [ 8 ]. For instance, ultrasound imaging can be used to examine and characterize tumors, whether benign or malignant.…”

“…On the other hand, DL-based QUS can directly classify the tissue state without needing a model to parameterize the signal. Therefore, QUS has recently evolved from model-based approaches [ 3 ], [ 4 ] to model-free, DL-based techniques [ 5 ], [ 6 ], [ 7 ], [ 8 ].…”