This paper studies a cooperative cognitive radio network where two primary users (PUs) exchange information with the help of a secondary user (SU) that is equipped with multiple antennas and in return, the SU superimposes its own messages along with the primary transmission. The fundamental problem in the considered network is the design of transmission strategies at the secondary node. It involves three basic elements: first, how to split the power for relaying the primary signals and for transmitting the secondary signals; second, what two-way relay strategy should be used to assist the bidirectional communication between the two PUs; third, how to jointly design the primary and secondary transmit precoders. This work aims to address this problem by proposing a transmission framework of maximizing the achievable rate of the SU while maintaining the rate requirements of the two PUs. Three well-known and practical two-way relay strategies are considered: amplify-and-forward (AF), bit level XOR based decode-and-forward (DF-XOR) and symbol level superposition coding based DF (DF-SUP). For each relay strategy, although the design problem is non-convex, we find the optimal solution by using certain transformation techniques and optimization tools such as semidefinite programming (SDP) and second-order cone programming (SOCP). Closed-form solutions are also obtained under certain conditions. Simulation results show that when the rate requirements of the two PUs are symmetric, by using the DF-XOR strategy and applying the proposed optimal precoding, the SU requires the least power for relaying and thus reserves the most power to transmit its own signal. In the asymmetric scenario, on the other hand, the DF-SUP strategy with the corresponding optimal precoding is the best.Comment: 31 pages 12 figures, Accepted by TS

No abstract

Liver X receptors (LXRs) are members of the superfamily of metabolic nuclear receptors, which play central roles in the regulation of cholesterol absorption, efflux, transportation and excretion and many other processes correlating with lipid metabolism. LXRs can also regulate inflammation in vitro and in vivo. Accumulating evidence demonstrates that LXR are involved in the metabolism and inflammation in human diseases. Nonalcoholic fatty liver disease (NAFLD) is classically associated with lipid metabolic disorders and inflammatory responses, especially in the nonalcoholic steatohepatitis (NASH) phase. The effects of LXRs on cholesterol metabolism and inflammation make them attractive as a potential target for the treatment of NAFLD. Since the ability to synthesize triglycerides may be protective in obesity and fatty liver, the hepatic lipogenesis by LXRs should not rule out the possibility of the use of LXRs in NAFLD.KEY WORDS: cholesterol metabolism, inflammatory response, liver X receptor, macrophage, nonalcoholic fatty liver disease.

The aim of this research is to automatically detect lumbar vertebras in MRI images with bounding boxes and their classes, which can assist clinicians with diagnoses based on large amounts of MRI slices. Vertebras are highly semblable in appearance, leading to a challenging automatic recognition. A novel detection algorithm is proposed in this paper based on deep learning. We apply a similarity function to train the convolutional network for lumbar spine detection. Instead of distinguishing vertebras using annotated lumbar images, our method compares similarities between vertebras using a beforehand lumbar image. In the convolutional neural network, a contrast object will not update during frames, which allows a fast speed and saves memory. Due to its distinctive shape, S1 is firstly detected and a rough region around it is extracted for searching for L1-L5. The results are evaluated with accuracy, precision, mean, and standard deviation (STD). Finally, our detection algorithm achieves the accuracy of 98.6% and the precision of 98.9%. Most failed results are involved with wrong S1 locations or missed L5. The study demonstrates that a lumbar detection network supported by deep learning can be trained successfully without annotated MRI images. It can be believed that our detection method will assist clinicians to raise working efficiency.