Carotid arteries vulnerable plaques are a crucial factor in the screening of atherosclerosis by ultrasound technique. However, manual plaque segmentation may be time-consuming and variable, moreover, the unstable plaques are contaminated by various noises such as artifacts and speckle noise. This paper proposes an automatic convolutional neural network (CNN) method for plaque segmentation in carotid ultrasound images using a small dataset. Firstly, a parallel network with three independent scale decoders is utilized as our base segmentation network, and pyramid dilated convolutions are used to enlarge receptive fields in three decoder sub-networks. Subsequently, the merged feature maps from the three decoders are rectified by the SENet. Thirdly, in the testing, the initial segmented plaque is refined by the maximal contour postprocessing method to obtain the final segmentation result. The dataset consists of 30 carotid ultrasound images with severe stenosis plaques from 30 patients. Test results show that the proposed method yields a Dice value of 0.820, IoU of 0.701, Accuracy of 0.969, and modified Hausdorff distance (MHD) of 1.43 by 10-fold cross-validation, it outperforms some CNN-based methods on these metrics. Additionally, we apply an ablation experiment to show the validity of each proposed module. Our method may be useful in actual applications for carotid unstable (easily ruptured or severe stenosis) plaques segmentation from ultrasound images.
Consistency training has proven to be an advanced semi-supervised framework and achieved promising results in medical image segmentation tasks through enforcing an invariance of the predictions over different views of the inputs. However, with the iterative updating of model parameters, the models would tend to reach a coupled state and eventually lose the ability to exploit unlabeled data. To address the issue, we present a novel semi-supervised segmentation model based on parameter decoupling strategy to encourage consistent predictions from diverse views. Specifically, we first adopt a two-branch network to simultaneously produce predictions for each image. During the training process, we decouple the two prediction branch parameters by quadratic cosine distance to construct different views in latent space. Based on this, the feature extractor is constrained to encourage the consistency of probability maps generated by classifiers under diversified features. In the overall training process, the parameters of feature extractor and classifiers are updated alternately by consistency regularization operation and decoupling operation to gradually improve the generalization performance of the model. Our method has achieved a competitive result over the state-of-the-art semi-supervised methods on the Atrial Segmentation Challenge dataset, demonstrating the effectiveness of our framework. Code is available at https://github.com/BX0903/PDC.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.