Wi-AM: Enabling Cross-Domain Gesture Recognition with Commodity Wi-Fi.

RF-based gesture recognition systems outperform computer vision-based systems in terms of user privacy. The integration of Wi-Fi sensing and deep learning has opened new application areas for intelligent multimedia technology. Although promising, existing systems have multiple limitations: (1) they only work well in a fixed domain; (2) when working in a new domain, they require the recollection of a large amount of data. These limitations either lead to a subpar cross-domain performance or require a huge amount of human effort, impeding their widespread adoption in practical scenarios. We propose Wi-AM, a privacy-preserving gesture recognition framework, to address the above limitations. Wi-AM can accurately recognize gestures in a new domain with only one sample. To remove irrelevant disturbances induced by interfering domain factors, we design a multi-domain adversarial scheme to reduce the differences in data distribution between different domains and extract the maximum amount of transferable features related to gestures. Moreover, to quickly adapt to an unseen domain with only a few samples, Wi-AM adopts a meta-learning framework to fine-tune the trained model into a new domain with a one-sample-per-gesture manner while achieving an accurate cross-domain performance. Extensive experiments in a real-world dataset demonstrate that Wi-AM can recognize gestures in an unseen domain with average accuracy of 82.13% and 86.76% for 1 and 3 data samples.

Retrospective study of the combined application of the blink reflex and lateral spread reflex in microvascular decompression of facial nerve surgery.

BACKGROUND: The blink reflex (BR) can be used as a routine monitoring method during facial nerve microvascular decompression. This study aimed to investigate whether the use of the BR in hemifacial spasm (HFS) surgery is comparable to that of the lateral spread reflex (LSR), and to explore its significance for guiding intraoperative neurophysiological monitoring (IONM). METHODS: Patients undergoing facial nerve microvascular decompression from 2016 to 2018 were included in the study. According to the results of IONM, the intraoperative monitoring items of the BR and two conventional facial nerve microvascular decompression procedures, namely the marginal mandibular branch LSR (MAR-LSR) and zygomatic branch LSR (ZYG-LSR), were compared. We mainly compared whether there were differences in the occurrence rate, disappearance rate, waveform, occurrence current, and prognosis of the three monitoring methods. RESULTS: The occurrence rate of the BR was lower than that of the MAR-LSR and ZYG-LSR, as well as the three combined detection groups. The disappearance rate of the BR was not different to that of the MAR-LSR, but higher than that of the ZYG-LSR group. In addition, the waveform of the BR showed differences from that of the MAR-LSR and ZYG-LSR. The incidence of postoperative residual symptoms in patients with any kind of reflex on the first day after surgery and the day of discharge was significantly higher than that of patients in which all three reflexes disappeared. CONCLUSIONS: Combined BR and LSR monitoring can reduce the occurrence of postoperative residual symptoms. We suggest that by increasing the use of BR examination during surgery, the integrity of the trigeminal nerve can be protected.

Alterations of Dynamic Regional Homogeneity in Trigeminal Neuralgia: A Resting-State fMRI Study.

Accumulating evidence from neuroimaging studies has supported that chronic pain could induce changes in brain function. However, few studies have focused on the dynamic regional homogeneity (dReHo) of trigeminal neuralgia (TN). In this study, twenty-eight TN patients and 28 healthy controls (HC) were included. Based on the resting-state fMRI (rsfMRI), we detected abnormalities in dReHo in the TN patients. Patients with TN had decreased dReHo in the left middle temporal gyrus, superior parietal lobule, and precentral gyrus, and increased dReHo in the thalamus. Furthermore, the increase in dReHo in the thalamus was positively correlated with duration of TN (r = 0.485, p = 0.012). These results provide compelling evidence for abnormal resting-state brain activity in TN and suggest that the duration of TN may play a critical role in brain function.

Autophagy inhibition promotes gambogic acid-induced suppression of growth and apoptosis in glioblastoma cells.

OBJECTIVE: To investigate the effects of gambogic acid (GA) on the growth of human malignant glioma cells. METHODS: U251MG and U87MG human glioma cell lines were treated with GA and growth and proliferation were investigated by MTT and colony formation assays. Cell apoptosis was analyzed by annexin V FITC/PI flow cytometry, mitochondrial membrane potential assays and DAPI nuclear staining. Monodansylcadaverine (MDC) staining and GFP-LC3 localisation were used to detect autophagy. Western blotting was used to investigate the molecular changes that occurred in the course of GA treatment. RESULTS: GA treatment significantly suppressed cell proliferation and colony formation, induced apoptosis in U251 and U87MG glioblastoma cells in a time- and dose-dependent manner. GA treatment also lead to the accumulation of monodansylcadaverine (MDC) in autophagic vacuoles, upregulated expressions of Atg5, Beclin 1 and LC3-II, and the increase of punctate fluorescent signals in glioblastoma cells pre-transfected with GFP-tagged LC3 plasmid. After the combination treatment of autophagy inhitors and GA, GA mediated growth inhibition and apoptotic cell death was further potentiated. CONCLUSION: Our results suggested that autophagic responses play roles as a self-protective mechanism in GA-treated glioblastoma cells, and autophagy inhibition could be a novel adjunctive strategy for enhancing chemotherapeutic effect of GA as an anti-malignant glioma agent.