Shadow Estimation for Ultrasound Images Using Auto-Encoding Structures and Synthetic Shadows

Abstract

Acoustic shadows are common artifacts in medical ultrasound imaging. The shadows are caused by objects that reflect ultrasound such as bones, and they are shown as dark areas in ultrasound images. Detecting such shadows is crucial for assessing the quality of images. This will be a pre-processing for further image processing or recognition aiming computer-aided diagnosis. In this paper, we propose an auto-encoding structure that estimates the shadowed areas and their intensities. The model once splits an input image into an estimated shadow image and an estimated shadow-free image through its encoder and decoder. Then, it combines them to reconstruct the input. By generating plausible synthetic shadows based on relatively coarse domain-specific knowledge on ultrasound images, we can train the model using unlabeled data. If pixel-level labels of the shadows are available, we also utilize them in a semi-supervised fashion. By experiments on ultrasound images for fetal heart diagnosis, we show that our method achieved 0.720 in the DICE score and outperformed conventional image processing methods and a segmentation method based on deep neural networks. The capability of the proposed method on estimating the intensities of shadows and the shadow-free images is also indicated through the experiments.

Integrated Analysis of Whole Genome and Epigenome Data Using Machine Learning Technology: Toward the Establishment of Precision Oncology

“…Currently, there are high expectations for the development of medical AI, and it is expected that AI technology will be actively introduced in actual clinical practice in the future. On the other hand, medical AI research for clinical applications is currently focused on medical image analysis (137)(138)(139)(140)(141)(142)(143)(144), and research on the introduction of AI to omics analysis such as whole genome analysis and epigenome analysis, as well as its clinical application, has not progressed sufficiently yet. In this regard, one of the problems associated with the widespread adoption of AI-based methodologies in omics analysis is that even though sequencing technology and other advanced analytics are increasingly being used in research and clinical practice, there is still a lot of confusion about the best protocols to adopt for analysis.…”

Towards Clinical Application of Artificial Intelligence in Ultrasound Imaging

“…Meng et al employed weakly supervised estimation of confidence maps using labels for each image with or without acoustic shadows [41,42]. Yasutomi et al proposed a semisupervised approach for integrating domain knowledge into a data-driven model using the pseudo-labeling of plausible synthetic shadows that were superimposed onto US imaging (Figure 1) [43]. The red areas represent the segmented acoustic shadows using the semi-supervised approach [43].…”

“…Yasutomi et al proposed a semisupervised approach for integrating domain knowledge into a data-driven model using the pseudo-labeling of plausible synthetic shadows that were superimposed onto US imaging (Figure 1) [43]. The red areas represent the segmented acoustic shadows using the semi-supervised approach [43]. (b) As a candidate for clinical application, examiners can evaluate whether the current acquired US imaging is suitable for diagnosis in real time.…”

Machine Learning Methods with Noisy, Incomplete or Small Datasets

“…The other papers that addressed medical applications are described as follows. In [4], Yasutomi et al (Japan) introduced a deep learning method based on an auto-encoder architecture to detect and remove shadow artifacts in ultrasound images. The model can be trained on unlabeled data (unsupervised) or with few pixel labels available (semi supervised).…”