Kisspeptin within the arcuate nucleus of the hypothalamus is a critical neuropeptide in the regulation of reproduction. Together with neurokinin B and dynorphin A, arcuate kisspeptin provides the oscillatory activity that drives the pulsatile secretion of gonadotrophin‐releasing hormone (GnRH), and therefore luteinising hormone (LH) pulses, and is considered to be a central component of the GnRH pulse generator. It is well established that the amygdala also exerts an influence over gonadotrophic hormone secretion and reproductive physiology. The discovery of kisspeptin and its receptor within the posterodorsal medial amygdala (MePD) and our recent finding showing that intra‐MePD administration of kisspeptin or a kisspeptin receptor antagonist results in increased LH secretion and decreased LH pulse frequency, respectively, suggests an important role for amygdala kisspeptin signalling in the regulation of the GnRH pulse generator. To further investigate the function of amygdala kisspeptin, the present study used an optogenetic approach to selectively stimulate MePD kisspeptin neurones and examine the effect on pulsatile LH secretion. MePD kisspeptin neurones in conscious Kiss1‐Cre mice were virally infected to express the channelrhodopsin 2 protein and selectively stimulated by light via a chronically implanted fibre optic cannula. Continuous stimulation using 5 Hz resulted in an increased LH pulse frequency, which was not observed at the lower stimulation frequencies of 0.5 and 2 Hz. In wild‐type animals, continuous stimulation at 5 Hz did not affect LH pulse frequency. These results demonstrate that selective activation of MePD Kiss1 neurones can modulate hypothalamic GnRH pulse generator frequency.

In this paper, we jointly consider communication, caching and computation in a multi-user cache-assisted mobile edge computing (MEC) system, consisting of one base station (BS) of caching and computing capabilities and multiple users with computation-intensive and latency-sensitive applications. We propose a joint caching and offloading mechanism which involves task uploading and executing for tasks with uncached computation results as well as computation result downloading for all tasks at the BS, and efficiently utilizes multi-user diversity and multicasting opportunities. Then, we formulate the average total energy minimization problem subject to the caching and deadline constraints to optimally allocate the storage resource at the BS for caching computation results as well as the uploading and downloading time durations. The problem is a challenging mixed discrete-continuous optimization problem. We show that strong duality holds, and obtain an optimal solution using a dual method. To reduce the computational complexity, we further propose a low-complexity suboptimal solution. Finally, numerical results show that the proposed suboptimal solution outperforms existing comparison schemes.

Acupuncture did not prevent PPOI and was not useful for treating PPOI once it had developed in this population.

Disordered emotion regulation may affect work efficiency, induce social disharmony, and even cause psychiatric diseases. Despite recent neurocomputing advances, whether positive and negative emotion networks can be voluntarily modulated is still unknown. In the present study, we addressed this question through multivariate voxel pattern analysis and real-time functional MRI neurofeedback (rtfMRI-nf). During a sustained emotion regulation task, participants' emotional states (positive or negative) were given to them as feedback. Participants were able to increase the percentage of positive emotional states, enhancing emotion regulation network activities. Participants showed an improvement on the positive subscale of positive and negative affect scale that came close to significance. Furthermore, the activation of several emotion-related brain regions, including insula, amygdala, anterior cingulate cortex, and dorsomedial prefrontal cortex, was also increased during rtfMRI-nf training. These findings suggest that humans are able to voluntarily modulate positive emotion networks, leading to exciting applications in the treatment of various neurological and psychiatric disorders.

There is now considerable evidence supporting the role of a subpopulation of neurons in the arcuate nucleus of the hypothalamus that coexpress kisspeptin, neurokinin B, and dynorphin (abbreviated as KNDy neurons) as the long sought-after gonadotropin-releasing hormone (GnRH) pulse generator. The “KNDy hypothesis” of pulse generation has largely been based on findings in rodents and ruminants, and there is considerably less information about the anatomical and functional organization of the KNDy subpopulation in the primate hypothalamus. In this review, we focus on the applicability of this hypothesis, and the roles of kisspeptin, neurokinin B, and dynorphin in reproduction, to humans and nonhuman primates, reviewing available data and pointing out important gaps in our current knowledge. With recent application of drugs that target KNDy peptides and their receptors to therapeutic treatments for reproductive disorders, it is imperative we fully understand the primate KNDy network and its role in the control of GnRH secretion, as well as species differences in this system that may exist between humans, nonhuman primates, and other mammals.

Progesterone can block the oestradiol-induced GnRH/LH surge and inhibit LH pulse frequency. Recent studies reported that progesterone prevented premature LH surges during ovarian hyperstimulation in women. As the most potent stimulator of GnRH/LH release, kisspeptin is believed to mediate the positive and negative feedback effects of oestradiol in the hypothalamic anteroventral periventricular (AVPV) and arcuate (ARC) nuclei, while the region-specific role of progesterone receptors in these nuclei remains unknown. This study examined the hypothesis that progesterone inhibits LH surge and pulsatile secretion via its receptor in the ARC and/or AVPV nuclei. Adult female rats received a single injection of pregnant mare serum gonadotropin followed by progesterone or vehicle. Progesterone administration resulted in a significant prolongation of the oestrous cycle and blockade of LH surge. However, microinjection of the progesterone receptor antagonist, RU486, into the AVPV reversed the prolonged cycle length and rescued the progesterone blockade LH surge, while RU486 into the ARC shortened LH pulse interval in the progesterone treated rats. These results demonstrated that progesterone’s inhibitory effect on the GnRH/LH surge and pulsatile secretion is mediated by its receptor in the kisspeptin enriched hypothalamic AVPV and ARC respectively, which are essential for progesterone regulation of oestrous cyclicity in rats.

We present an automatic method based on transfer learning for the identification of dry age-related macular degeneration (AMD) and diabetic macular edema (DME) from retinal optical coherence tomography (OCT) images. The algorithm aims to improve the classification performance of retinal OCT images and shorten the training time. Firstly, we remove the last several layers from the pre-trained Inception V3 model and regard the remaining part as a fixed feature extractor. Then, the features are used as input of a convolutional neural network (CNN) designed to learn the feature space shifts. The experimental results on two different retinal OCT images datasets demonstrate the effectiveness of the proposed method.

Finetuning pre-trained deep neural networks (DNN) delicately designed for large-scale natural images may not be suitable for medical images due to the intrinsic difference between the datasets. We propose a strategy to modify DNNs, which improves their performance on retinal optical coherence tomography (OCT) images. Deep features of pre-trained DNN are high-level features of natural images. These features harm the training of transfer learning. Our strategy is to remove some deep convolutional layers of the state-of-the-art pre-trained networks: GoogLeNet, ResNet and DenseNet. We try to find the optimized deep neural networks on small-scale and large-scale OCT datasets, respectively, in our experiments. Results show that optimized deep neural networks not only reduce computational burden, but also improve classification accuracy.