The patient with ischemic stroke can benefit most from the earliest possible definitive diagnosis. While a quantitative evaluation of the stroke lesions on the magnetic resonance images (MRIs) is effective in clinical diagnosis, manually segmenting the stroke lesions is commonly used, which is, however, a tedious and time-consuming task. Therefore, how to segment the stroke lesions in a fully automated manner has recently extracted extensive attentions. Considering that the clinically acquired MRIs usually have thick slices, we propose a 2D-slice-based segmentation method. In particular, we use multi-spectral MRIs, i.e., diffusion weighted image, apparent diffusion coefficient, and T2-weighted image, as input, and propose a residual-structured fully convolutional network (Res-FCN). The proposed Res-FCN is trained and evaluated on a large data set with 212 clinically acquired MRIs, which achieves a mean dice coefficient of 0.645 with a mean number of false negative lesions of 1.515 per subject. The proposed Res-FCN is further evaluated on a public data set, i.e., ISLES2015-SISS, which presents a very competitive result among all 2D-slice-based segmentation methods.

Acute ischemic stroke (AIS) has been a common threat to human health and may lead to severe outcomes without proper and prompt treatment. To precisely diagnose AIS, it is of paramount importance to quantitatively evaluate the AIS lesions. By adopting a convolutional neural network (CNN), many automatic methods for ischemic stroke lesion segmentation on magnetic resonance imaging (MRI) have been proposed. However, most CNN-based methods should be trained on a large amount of fully labeled subjects, and the label annotation is a labor-intensive and time-consuming task. Therefore, in this paper, we propose to use a mixture of many weakly labeled and a few fully labeled subjects to relieve the thirst of fully labeled subjects. In particular, a multifeature map fusion network (MFMF-Network) with two branches is proposed, where hundreds of weakly labeled subjects are used to train the classification branch, and several fully labeled subjects are adopted to tune the segmentation branch. By training on 398 weakly labeled and 5 fully labeled subjects, the proposed method is able to achieve a mean dice coefficient of on a test set with 179 subjects. The lesion-wise and subject-wise metrics are also evaluated, where a lesion-wise F1 score of 0.886 and a subject-wise detection rate of 1 are achieved.

Accurate inferior alveolar nerve (IAN) canal segmentation has been considered a crucial task in dentistry. Failing to accurately identify the position of the IAN canal may lead to nerve injury during dental procedures. While IAN canals can be detected from dental cone beam computed tomography, they are usually difficult for dentists to precisely identify as the canals are thin, small, and span across many slices. This paper focuses on improving accuracy in segmenting the IAN canals. By integrating our proposed frequency-domain attention mechanism in UNet, the proposed frequency attention UNet (FAUNet) is able to achieve $75.55\%$ and $81.35\%$ in the Dice and surface Dice coefficients, respectively, which are much higher than other competitive methods, by adding only 224 parameters to the classical UNet. Compared to the classical UNet, our proposed FAUNet achieves a $2.39\%$ and $2.82\%$ gain in the Dice coefficient and the surface Dice coefficient, respectively. The potential advantage of developing attention in the frequency domain is also discussed, which revealed that the frequency-domain attention mechanisms can achieve better performance than their spatial-domain counterparts.

Vehicular communication in high mobility environments has been widely explored in recent years. However, due to the existence of carrier frequency offset (CFO) and dynamic channel, the performance of vehicular communication systems over fast time-varying scenes drops severely. To address this problem, in this paper, we propose a simple and effective CRS-based joint CFO and channel estimation baseline method using deep learning (DL) for orthogonal frequency division multiplexing (OFDM) systems. Concretely, we construct a joint neural network (NN) architecture consisting of a CFO estimation network (CFOENet) based on fully connected layers and a channel estimation network (CENet) composed of convolutional layers. The proposed NN architecture can fully exploit the correlation of the cell reference signal (CRS), while learning the CFO characteristics and channel state information (CSI) changes simultaneously, which highly improves the pilot usage efficiency. We conduct adequate simulation experiments, and the results demonstrate that the proposed DL-based scheme can achieve better performance in terms of CFO estimation, channel estimation and overall system performance than conventional methods, while our method has stronger robustness and generalization ability under various channel conditions. The proposed joint CFO and channel estimation scheme has great potential in the field of the Internet of Vehicles.

Ultra-Reliable Low-Latency Communication (URLLC) has imposed significant challenges on the Internet of Things (IoT) networks due to its stringent Quality of Service (QoS) requirements. Conventionally, to support the latency and reliability requirements, the IoT devices have to leverage the transmission power, which significantly shortened the battery life. As a promising candidate for the future mobile communication systems, reconfigurable intelligent surface (RIS) is preferrable to be deployed in the IoT networks to reduce the power consumption, thanks to its ability in reconfiguring the propagation environment. In this paper, we focus on the design of an aerial RIS aided URLLC system where the RIS is deployed on an unmanned aerial vehicle (UAV). The system operates on the time division multiple access (TDMA) protocol, and the transmission power minimization problem is investigated. A low-complexity algorithm is proposed to jointly optimize the RIS phase shifts, transmission time duration, and the UAV location. Simulation results demonstrate that by deploying aerial RIS in the IoT network, the transmission power at the IoT device can be significantly reduced. The effect of the TDMA frame length is also discussed. When the TDMA frame length is large, the optimal transmission time duration could be much shorter than equal allocation, which implies that our proposed design is able to simultaneously reduce both the average power consumption and the transmission latency.