Unlike daily routine images, ultrasound images are usually monochrome and low-resolution. In ultrasound images, the cancer regions are usually blurred, vague margin and irregular in shape. Moreover, the features of cancer region are very similar to normal or benign tissues. Therefore, training ultrasound images with original Convolutional Neural Network (CNN) directly is not satisfactory. In our study, inspired by state-of-the-art object detection network Faster R-CNN, we develop a detector which is more suitable for thyroid papillary carcinoma detection in ultrasound images. In order to improve the accuracy of the detection, we add a spatial constrained layer to CNN so that the detector can extract the features of surrounding region in which the cancer regions are residing. In addition, by concatenating the shallow and deep layers of the CNN, the detector can detect blurrier or smaller cancer regions. The experiments demonstrate that the potential of this new methodology can reduce the workload for pathologists and increase the objectivity of diagnoses. We find that 93:5% of papillary thyroid carcinoma regions could be detected automatically while 81:5% of benign and normal tissue could be excluded without the use of any additional immunohistochemical markers or human intervention.

An improved faster region-based convolutional neural network (R-CNN) [same object retrieval (SOR) faster R-CNN] is proposed to retrieve the same object in different scenes with few training samples. By concatenating the feature maps of shallow and deep convolutional layers, the ability of Regions of Interest (RoI) pooling to extract more detailed features is improved. In the training process, a pretrained CNN model is fine-tuned using a query image data set, so that the confidence score can identify an object proposal to the object level rather than the classification level. In the query process, we first select the ten images for which the object proposals have the closest confidence scores to the query object proposal. Then, the image for which the detected object proposal has the minimum cosine distance to the query object proposal is considered as the query result. The proposed SOR faster R-CNN is applied to our Coke cans data set and three public image data sets, i.e., Oxford Buildings 5k, Paris Buildings 6k, and INS 13. The experimental results confirm that SOR faster R-CNN has better identification performance than fine-tuned faster R-CNN. Moreover, SOR faster R-CNN achieves much higher accuracy for detecting low-resolution images than the fine-tuned faster R-CNN on the Coke cans (0.094 mAP higher), Oxford Buildings (0.043 mAP higher), Paris Buildings (0.078 mAP higher), and INS 13 (0.013 mAP higher) data sets.

Dropout is one of the most widely used methods to avoid overfitting neural networks. However, it rigidly and randomly activates neurons according to a fixed probability, which is not consistent with the activation mode of neurons in the human cerebral cortex. Inspired by gene theory and the activation mechanism of brain neurons, we propose a more intelligent adaptive dropout, in which a variational self-encoder (VAE) overlaps to an existing neural network to regularize its hidden neurons by adaptively setting activities to zero. Through alternating iterative training, the discarding probability of each hidden neuron can be learned according to the weights and thus effectively avoid the shortcomings of the standard dropout method. The experimental results in multiple data sets illustrate that this method can better suppress overfitting in various neural networks than can the standard dropout. Additionally, this adaptive dropout technique can reduce the number of neurons and improve training efficiency.

The application of batteries has become more and more extensive, and the heat dissipation problem cannot be ignored. Oscillating Heat Pipe (OHP) is a good means of heat dissipation. In this paper, the methods to improve the energy conversion and flow thermal performance of micro-channel OHP are studied and summarized. The working principle, heat transfer mechanism, advantages and applications of PHP are also introduced in detail in this study. Proper adjustment of the micro-channel layout can increase the heat transfer limit of PHP by 44%. The thermal resistance of two-diameter channel PHP is 45% lower than that of conventional PHP. The thermal resistance of PHP under uneven heating can be reduced to 50% of the original. PHP pulse heating can alleviate the phenomenon of dryness. Different working fluids have different effects on PHP. The use of graphene nano-fluids as the work medium can reduce the thermal resistance of PHP by 83.6%. The work medium obtained by the mixture of different fluids has the potential to compensate for the defects while inheriting the advantages of a single fluid.

Short‐term voltage stability (STVS) assessment is critical for ensuring the operational security of modern industrial internet‐based power systems. For data‐driven STVS evaluation approaches, effectively leveraging both time‐series data and topological structure of complex industrial power networks to extract critical spatial‐temporal features remains a challenge. This paper introduces a novel spatial‐temporal feature learning framework, termed spatial‐temporal graph attention network (STGAT), which integrates graph attention network (GAT) and bidirectional gated recurrent unit (BiGRU). In the framework, channel attention mechanism (CAM) is incorporated into the GAT to enhance spatial representation, while temporal attention mechanism is applied to the BiGRU to capture essential temporal features. By considering highly representative spatial‐temporal correlations of power system dynamics, the recommended STGAT model delivers a fast, accurate, and robust classification framework for STVS assessment. Extensive testing on the IEEE 39‐bus system validates the feasibility and preeminence of the recommended STGAT method compared to existing models, ensuring its suitability for online STVS assessment in industrial environments.