Medical image segmentation is a key topic in image processing and computer vision. Existing literature mainly focuses on single-organ segmentation. However, since maximizing the concentration of radiotherapy drugs in the target area with protecting the surrounding organs is essential for making effective radiotherapy plan, multiorgan segmentation has won more and more attention. An improved Mask R-CNN (region-based convolutional neural network) model is proposed for multiorgan segmentation to aid esophageal radiation treatment. Due to the fact that organ boundaries may be fuzzy and organ shapes are various, original Mask R-CNN works well on natural image segmentation while leaves something to be desired on the multiorgan segmentation task. Addressing it, the advantages of this method are threefold: (1) a ROI (region of interest) generation method is presented in the RPN (region proposal network) which is able to utilize multiscale semantic features. (2) A prebackground classification subnetwork is integrated to the original mask generation branch to improve the precision of multiorgan segmentation. (3) 4341 CT images of 44 patients are collected and annotated to evaluate the proposed method. Additionally, extensive experiments on the collected dataset demonstrate that the proposed method can segment the heart, right lung, left lung, planning target volume (PTV), and clinical target volume (CTV) accurately and efficiently. Specifically, less than 5% of the cases were missed detection or false detection on the test set, which shows a great potential for real clinical usage.
The objective of facial attribute recognition is to predict a set of labels for each face image. Similar to general multi-label image classification task, facial attribute recognition suffers from several typical problems: discriminative feature learning, handling the correlations among attributes, imbalanced training data, etc. In this paper, we propose a unified facial attribute recognition solution via feature decoupling and graph convolutional networks (GCN): (1) We utilize the orthonormal regularizer to constrain that each dimension of the general facial representation represents a certain visual pattern. Meanwhile, a learnable matrix is unveiled, which can convert general facial feature into attribute-specific representation through modeling the importance of different visual patterns. (2) To handle the correlations between facial attributes, we build a GCN to capture the label dependencies and map the nodes in the proposed GCN to a set of inter-dependent attribute classifiers. Besides, we normalize the weights in all classifiers to alleviate the influence of data imbalance. We have conducted extensive experiments on two benchmarks, and both qualitative and quantitative evaluation results have demonstrated the effectiveness of the proposed method.
1 October 2020 in Mathematical Problems in Engineering
Precipitation nowcasting plays a key role in land security and emergency management of natural calamities. A majority of existing deep learning-based techniques realize precipitation nowcasting by learning a deep nonlinear function from a single information source, e.g., weather radar. In this study, we propose a novel multimodal semisupervised deep graph learning framework for precipitation nowcasting. Unlike existing studies, different modalities of observation data (including both meteorological and nonmeteorological data) are modeled jointly, thereby benefiting each other. All information is converted into image structures, next, precipitation nowcasting is deemed as a computer vision task to be optimized. To handle areas with unavailable precipitation, we convert all observation information into a graph structure and introduce a semisupervised graph convolutional network with a sequence connect architecture to learn the features of all local areas. With the learned features, precipitation is predicted through a multilayer fully connected regression network. Experiments on real datasets confirm the effectiveness of the proposed method.