Search

The clinical diagnosis of gastrointestinal stromal tumors (GISTs) requires time-consuming tumor localization by physicians, while automated detection of GIST can help physicians develop timely treatment plans. Existing GIST detection methods based on fully supervised deep learning require a large amount of labeled data for the model training, but the acquisition of labeled data is often time-consuming and labor-intensive, hindering the optimization of the model. However, the semi-supervised learning method can perform better than the fully supervised learning method with only a small amount of labeled data because of the full use of unlabeled data, which effectively compensates for the lack of labeled data. Therefore, we propose a semi-supervised gastrointestinal stromal tumor (GIST) detection method based on self-training using the new selection criterion to guarantee the quality of pseudo-labels and adding the pseudo-labeled data to the training set together with the labeled data after linear mixing. In addition, we introduce the improved Faster RCNN with the multiscale module and the feature enhancement module (FEM) for semi-supervised GIST detection. The multiscale module and the FEM can better fit the characteristics of GIST and obtain better detection results. The experiment results showed that our approach achieved the best performance on our GIST image dataset with the joint optimization of the self-training framework, the multiscale module, and the FEM. [ABSTRACT FROM AUTHOR]

Compressional and shear wave velocities (Vp and Vs, respectively) are important elastic parameters to predict reservoir parameters, such as lithology and hydrocarbons. Due to acquisition technologies and economy, the shear wave velocity is generally lacking. Over the last few years, some researchers proposed deep learning algorithms to predict the shear wave velocity using conventional logging data. However, these algorithms focus either on spatial feature extraction for different physical properties of rocks or on sequential feature extraction in the depth direction of rocks. Only focusing on feature extraction in a direction of rocks might lead to a decrease in prediction accuracy. Therefore, we propose a hybrid network of a two-dimensional convolutional neural network and the gated recurrent unit (2DCNN-GRU), which can establish more complex nonlinear relationships between the input and output data based on the spatial features extracted by 2DCNN and the sequential features extracted by GRU. In this study, two cases are used to validate the reliability and prediction accuracy of the proposed network. Comparing the prediction results of 2DCNN, GRU, and the proposed network, the proposed network shows better performance. Meanwhile, for improving the prediction accuracy of the proposed network, the relationship is analyzed between the prediction accuracy of the proposed network and the length of the input sample. [ABSTRACT FROM AUTHOR]

Computational intelligence has been widely used in medical information processing. The deep learning methods, especially, have many successful applications in medical image analysis. In this paper, we proposed an end-to-end medical lesion segmentation framework based on convolutional neural networks with a dual attention mechanism, which integrates both fully and weakly supervised segmentation. The weakly supervised segmentation module achieves accurate lesion segmentation by using bounding-box labels of lesion areas, which solves the problem of the high cost of pixel-level labels with lesions in the medical images. In addition, a dual attention mechanism is introduced to enhance the network's ability for visual feature learning. The dual attention mechanism (channel and spatial attention) can help the network pay attention to feature extraction from important regions. Compared with the current mainstream method of weakly supervised segmentation using pseudo labels, it can greatly reduce the gaps between ground-truth labels and pseudo labels. The final experimental results show that our proposed framework achieved more competitive performances on oral lesion dataset, and our framework further extended to dermatological lesion segmentation. [ABSTRACT FROM AUTHOR]