Shadow Estimation for Ultrasound Images Using Auto-Encoding Structures and Synthetic Shadows
Autor: | Reina Komatsu, Akihiko Sekizawa, Kanto Shozu, Suguru Yasutomi, Tatsuya Arakaki, Ai Dozen, Ryuji Hamamoto, Hidenori Machino, Ken Asada, Syuzo Kaneko, Ryu Matsuoka, Masaaki Komatsu, Akira Sakai |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2021 |
Předmět: |
semi-supervised learning
genetic structures Computer science ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Image processing Semi-supervised learning macromolecular substances lcsh:Technology 030218 nuclear medicine & medical imaging lcsh:Chemistry 03 medical and health sciences 0302 clinical medicine Encoding (memory) Shadow auto-encoders General Materials Science Segmentation Computer vision shadow estimation ultrasound images Instrumentation lcsh:QH301-705.5 ComputingMethodologies_COMPUTERGRAPHICS Fluid Flow and Transfer Processes integumentary system business.industry lcsh:T Process Chemistry and Technology Deep learning General Engineering deep learning respiratory system Acoustic shadow shadow detection lcsh:QC1-999 Computer Science Applications respiratory tract diseases lcsh:Biology (General) lcsh:QD1-999 lcsh:TA1-2040 Artificial intelligence business lcsh:Engineering (General). Civil engineering (General) Encoder 030217 neurology & neurosurgery lcsh:Physics |
Zdroj: | Applied Sciences Volume 11 Issue 3 Applied Sciences, Vol 11, Iss 1127, p 1127 (2021) |
ISSN: | 2076-3417 |
DOI: | 10.3390/app11031127 |
Popis: | 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. |
Databáze: | OpenAIRE |
Externí odkaz: |