Iterative Multi-domain Regularized Deep Learning for Anatomical Structure Detection and Segmentation from Ultrasound Images
This addresses the need for accurate clinical diagnosis and biometric measurements from ultrasound images, but appears incremental as it builds on existing deep learning techniques with iterative and multi-domain refinements.
The paper tackled the problem of detecting and segmenting anatomical structures in ultrasound images, which is challenging due to fuzzy borders and artifacts, and reported that their method achieved superior accuracy, outperforming other methods by a significant margin and demonstrating competitive capability compared to human performance.
Accurate detection and segmentation of anatomical structures from ultrasound images are crucial for clinical diagnosis and biometric measurements. Although ultrasound imaging has been widely used with superiorities such as low cost and portability, the fuzzy border definition and existence of abounding artifacts pose great challenges for automatically detecting and segmenting the complex anatomical structures. In this paper, we propose a multi-domain regularized deep learning method to address this challenging problem. By leveraging the transfer learning from cross domains, the feature representations are effectively enhanced. The results are further improved by the iterative refinement. Moreover, our method is quite efficient by taking advantage of a fully convolutional network, which is formulated as an end-to-end learning framework of detection and segmentation. Extensive experimental results on a large-scale database corroborated that our method achieved a superior detection and segmentation accuracy, outperforming other methods by a significant margin and demonstrating competitive capability even compared to human performance.