Machine Learning Assisted Doppler Features for Enhancing Thyroid Cancer Diagnosis

Abstract

Background This pilot study aims at exploiting machine learning techniques to extract color Doppler ultrasound (CDUS) features and to build an artificial neural network (ANN) model based on these CDUS features for improving the diagnostic performance of thyroid cancer classification. Methods A total of 674 patients with 712 thyroid nodules (TNs) (512 from internal dataset and 200 from external dataset) were randomly selected in this retrospective study. We used ANN to build a model (TDUS‐Net) for classifying malignant and benign TNs using both the automatically extracted quantitative CDUS features (whole ratio, intranodular ratio, peripheral ratio, and number of vessels) and gray‐scale ultrasound (US) features defined by the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI‐RADS). Then, we compared the diagnostic performance of the model, the performance of another ANN model based on the gray‐scale US features alone (TUS‐Net), and that of radiologists. Results The TDUS‐Net (0.898, 95% CI: 0.868–0.922) achieved a higher area under the curve (AUC) than that of TUS‐Net (0.881, 95% CI: 0.850–0.908) in the internal tests. Compared with radiologists, TDUS‐Net (AUC: 0.925, 95% CI: 0.880–0.958) performed better than radiologists (AUC: 0.810, 95% CI: 0.749–0.862) in the external tests. Conclusions Applying a machine learning model by combining both gray‐scale US features and CDUS features can achieve comparable or even higher performance than radiologists in classifying TNs.

Recent Advances in Ultrasound and Photoacoustic Analysis for Thyroid Cancer Diagnosis

“…[69] Recently, Zhu et al demonstrated a machine learning-based technique for enhancing diagnostic performance using US Doppler data acquired from 712 thyroid nodules (356 malignant and 356 benign). [70] They designed an artificial neural network model (TDUS-Net) by extracting features from US Doppler images (whole ratio, intranodular ratio, peripheral ratio, and the number of vessels) and US features defined by TI-RADS. For comparison, an additional artificial neural network model (TUS-Net) based on the US features alone was also designed.…”