Graph Convolutional Adversarial Networks for Spatiotemporal Anomaly Detection.

Traffic anomalies, such as traffic accidents and unexpected crowd gathering, may endanger public safety if not handled timely. Detecting traffic anomalies in their early stage can benefit citizens' quality of life and city planning. However, traffic anomaly detection faces two main challenges. First, it is challenging to model traffic dynamics due to the complex spatiotemporal characteristics of traffic data. Second, the criteria of traffic anomalies may vary with locations and times. In this article, we propose a spatiotemporal graph convolutional adversarial network (STGAN) to address these above challenges. More specifically, we devise a spatiotemporal generator to predict the normal traffic dynamics and a spatiotemporal discriminator to determine whether an input sequence is real or not. There are high correlations between neighboring data points in the spatial and temporal dimensions. Therefore, we propose a recent module and leverage graph convolutional gated recurrent unit (GCGRU) to help the generator and discriminator learn the spatiotemporal features of traffic dynamics and traffic anomalies, respectively. After adversarial training, the generator and discriminator can be used as detectors independently, where the generator models the normal traffic dynamics patterns and the discriminator provides detection criteria varying with spatiotemporal features. We then design a novel anomaly score combining the abilities of two detectors, which considers the misleading of unpredictable traffic dynamics to the discriminator. We evaluate our method on two real-world datasets from New York City and California. The experimental results show that the proposed method detects various traffic anomalies effectively and outperforms the state-of-the-art methods. Furthermore, the devised anomaly score achieves more robust detection performances than the general score.

[Melatonin receptor 2 mediated spinal dorsal horn astrocytes IL-17 inhibition contributes to electroacupuncture analgesia in neuropathic pain rats].

OBJECTIVE: To investigate the effect of electroacupuncture (EA) on pain behaviors and expression of spinal dorsal horn melatonin receptor 2 (MT2) and interleukin-17 (IL-17) in neuropathic pain rats, so as to explore its mechanism underlying pain relief. METHODS: The present study includes 3 parts. In the first part, eighteen male SD rats were randomly divided into 3 groups: sham operation, model and EA groups, with 6 rats in each group. The neuropathic pain model was established by chronic constriction injury (CCI) of the right sciatic nerve. On the 7th day following modeling, EA was applied to the right "Zusanli" (ST36) and "Sanyinjiao" (SP6) (1 mA,2 Hz/100 Hz) for 30 min. The mechanical pain threshold(MWT) and thermal pain thre-shold(TPT) of the affected limb were detected before modeling, 7 days following modeling and 60 min after EA. The expression of MT2 in spinal dorsal horn was detected by Western blot. The contents of melatonin （Mel） and IL-17 in the spinal dorsal horn were determined by ELISA. The expression of glial fibrillary acidic protein (GFAP) in the spinal dorsal horn was determined by Western blot and immunohistochemistry. In the second part, 30 rats were divided into 5 groups: sham operation, model, EA, MT2 antagonist (4-P-PDOT), and dimethyl sulfoxide (DMSO) groups, with 6 rats in each group. Rats of the 4-P-PDOT and DMSO groups were intrathecal injection with 10 µL MT2 antagonist 4-P-PDOT (100 µg) and equivalent DMSO 30 min before EA. The MWT and TPT of affected limb were detected. The GFAP expression and IL-17 content in the spinal dorsal horn was detected by Western blot， immunohistochemistry and ELISA, respectively. In the third part, 30 rats were randomly divided into 5 groups: sham operation, model, EA, recombinant IL-17, and normal saline groups, with 6 rats in each group. The recombinant IL-17 protein (100 ng, 10 µL) and the same amount of 0.9% sodium chloride solution were intrathecal injection into the rats of the recombinant IL-17 group and the normal saline group 30 min before the EA. The MWT and TPT of affected limb were measured. RESULTS: On the 7th day after modeling, the MWT of rats in the model group and the EA group were significantly higher, while TPT were lower than those before the modeling (P<0.05). At 60 min after EA, compared with the model group, the MWT and TPT of the EA group reversed significantly (P<0.05). The levels of GFAP and IL-17 were significantly increased, while the levels of Mel and MT2 were significantly decreased in the model group than in the sham operation group (P<0.05), and those were considerably reversed in the EA group than in the model group (P<0.05). Compared with the EA and DMSO groups, the MWT in the 4-P-PDOT group were significantly increased, while TPT were decreased (P<0.05), and the contents of GFAP and IL-17 were significantly increased (P<0.05). Compared to the EA and normal saline groups, MWT of the rats in the recombinant IL-17 group were significantly increased, while TPT decreased (P<0.05). CONCLUSION: EA of ST36 and SP6 can alleviate neuropathic pain in CCI rats, which is closely related to its effect in inhibiting the release of IL-17 from astrocytes mediated by MT2.