Lenghu on the Tibetan Plateau as an astronomical observing site.

On Earth's surface, there are only a handful of high-quality astronomical sites that meet the requirements for very large next-generation facilities. In the context of scientific opportunities in time-domain astronomy, a good site on the Tibetan Plateau will bridge the longitudinal gap between the known best sites1,2 (all in the Western Hemisphere). The Tibetan Plateau is the highest plateau on Earth, with an average elevation of over 4,000 metres, and thus potentially provides very good opportunities for astronomy and particle astrophysics3-5. Here we report the results of three years of monitoring of testing an area at a local summit on Saishiteng Mountain near Lenghu Town in Qinghai Province. The altitudes of the potential locations are between 4,200 and 4,500 metres. An area of over 100,000 square kilometres surrounding Lenghu Town has a lower altitude of below 3,000 metres, with an extremely arid climate and unusually clear local sky (day and night)6. Of the nights at the site, 70 per cent have clear, photometric conditions, with a median seeing of 0.75 arcseconds. The median night temperature variation is only 2.4 degrees Celsius, indicating very stable local surface air. The precipitable water vapour is lower than 2 millimetres for 55 per cent of the night.

Effectiveness evaluation of flipped classroom in emergency medicine online teaching for medical undergraduates.

INTRODUCTION: Due to policy changes in the context of COVID-19 pandemic, online teaching has become the main form of class in many Chinese universities. Flipped classroom has been widely used in other disciplines, but there is a dearth of evidence available about the use in online teaching of emergency medicine. This study aimed to develop a flipped classroom for online emergency medicine teaching and evaluate its effectiveness by comparing it with traditional lecture-based online teaching. METHODS: A total of 62 clinical medical undergraduates from Jinan University participated in this study from September to December in 2022. An online flipped classroom approach was developed (FC group, n = 31). Traditional lecture-based online teaching was applied as a contrast (LBT group, n = 31). The undergraduates completed examinations and questionnaires at the end of the course. A course experience questionnaire and course examination score were used to evaluate the effectiveness of the flipped classroom approach. RESULTS: Regarding the five dimensions of the course experience questionnaire, the scores for good teaching (3.47 ± 0.50 vs. 2.34 ± 0.48, p < .001), appropriate assessment (3.31 ± 0.68 vs. 2.95 ± 0.71, p = .043) and generic skills (3.16 ± 0.60 vs. 2.72 ± 0.39, p < .001) were higher for the FC group than for the LBT group. There was no significant difference between the two groups in clear goals and standards, and appropriate workload. The undergraduates in the FC group showed significantly higher overall satisfaction than those in the LBT group (3.52 ± 0.1.03 vs. 2.87 ± 0.92, p = .012). The examination scores (77.936 ± 11.573 vs. 70.484 ± 7.434, p < .001), especially the scores for questions related to case analysis (33.032 ± 5.363 vs. 26.968 ± 7.657, p < .001), were significantly higher in the FC group than in the LBT group. CONCLUSIONS: The flipped classroom for online teaching was efficient in improving undergraduates' emergency medical academic performance and promoting the development of clinical case analysis ability. These findings provide an alternative flipped classroom approach for online teaching of emergency medicine.

Training much deeper spiking neural networks with a small number of time-steps.

Spiking Neural Network (SNN) is a promising energy-efficient neural architecture when implemented on neuromorphic hardware. The Artificial Neural Network (ANN) to SNN conversion method, which is the most effective SNN training method, has successfully converted moderately deep ANNs to SNNs with satisfactory performance. However, this method requires a large number of time-steps, which hurts the energy efficiency of SNNs. How to effectively covert a very deep ANN (e.g., more than 100 layers) to an SNN with a small number of time-steps remains a difficult task. To tackle this challenge, this paper makes the first attempt to propose a novel error analysis framework that takes both the "quantization error" and the "deviation error" into account, which comes from the discretization of SNN dynamicsthe neuron's coding scheme and the inconstant input currents at intermediate layers, respectively. Particularly, our theories reveal that the "deviation error" depends on both the spike threshold and the input variance. Based on our theoretical analysis, we further propose the Threshold Tuning and Residual Block Restructuring (TTRBR) method that can convert very deep ANNs (>100 layers) to SNNs with negligible accuracy degradation while requiring only a small number of time-steps. With very deep networks, our TTRBR method achieves state-of-the-art (SOTA) performance on the CIFAR-10, CIFAR-100, and ImageNet classification tasks.

Camrelizumab plus apatinib as second-line treatment for advanced oesophageal squamous cell carcinoma (CAP 02): a single-arm, open-label, phase 2 trial.

BACKGROUND: Camrelizumab, an anti-PD-1 antibody, has shown moderate efficacy in oesophageal squamous cell carcinoma. Apatinib, a selective inhibitor of VEGFR2, has a synergistic effect with immunotherapy. We aimed to assess the combination of camrelizumab and apatinib as second-line treatment for advanced oesophageal squamous cell carcinoma. METHODS: This single-arm, open-label, phase 2 study was conducted at eight centres in China. Eligible patients were aged 18-75 years, with an Eastern Cooperative Oncology Group performance status of 0 or 1, who had unresectable locally advanced, locally recurrent, or metastatic oesophageal squamous cell carcinoma, and had progressed after or were intolerant to first-line chemotherapy. Patients received intravenous camrelizumab 200 mg once every 2 weeks plus oral apatinib 250 mg once daily for a 28-day cycle until disease progression, unacceptable adverse events, or withdrawal of consent. The primary endpoint was investigator-assessed confirmed objective response rate. Efficacy was analysed in patients who had received at least one dose of study drug, and safety was analysed in patients who received the study drug and had at least one post-baseline safety assessment. The study of this cohort is complete and this trial is registered with ClinicalTrials.gov, number NCT03736863. FINDINGS: Between Dec 5, 2019, and Feb 10, 2021, 52 patients were enrolled and included in analyses. At data cutoff (June 20, 2021), median follow-up was 7·5 months (IQR 4·0-11·2). 18 (34·6%, [95% CI 22·0-49·1]) of 52 patients had a confirmed objective response. 23 (44%) of 52 patients had grade 3 or worse treatment-related adverse events. The most common grade 3 or worse treatment-related adverse events were increased aspartate aminotransferase (10 [19%]), increased gamma-glutamyltransferase (10 [19%]), and increased alanine aminotransferase (five [10%]). No treatment-related deaths occurred. INTERPRETATION: Camrelizumab combined with apatinib showed promising activity and manageable toxicity, and might be a potential second-line treatment option for patients with advanced oesophageal squamous cell carcinoma. Another cohort of this study, enrolling patients previously treated with first-line immunotherapy, is ongoing. FUNDING: Jiangsu Hengrui Pharmaceuticals.

Micro-fabricated perforated polymer devices for long-term drug delivery.

Fabrication techniques have been developed to produce a perforated polymer microtube as a drug delivery device. The technique consists of first forming a silicon platform with trenches and alignment marks to hold the tubes for subsequent processing. Photolithography and reactive ion etching with an inductively coupled plasma source were used to fabricate micro holes on the surface of polyimide tubes. Several materials have been used to form the etching mask, including titanium film deposited by e-beam evaporation and SiO(2) and SiN(x) films deposited by high-density plasma chemical vapor deposition (HDPCVD). Three equidistant holes of 20 µm in diameter were fabricated on polyimide tubes (I.D. = 125 µm). The perforated tubes were loaded with ethinyl estradiol and tested for drug release in phosphate buffered saline (pH = 7.1) at 37°C. Zero order release was observed over a period of 30 days with a potential to be extended to 4 years.

Development and characterization of a scalable microperforated device capable of long-term zero order drug release.

A drug delivery system that consists of microperforated polyimide microtubes was developed and characterized. Two groups of polyimide tubes were used. One set consisted of microtubes (I.D. = 125 microm) with 32.9 +/- 1.7 microm size holes. The second set consisted of larger tubes (I.D. = 1000 microm) with 362-542 microm holes. The number of holes was varied between 1 and 3. The small tubes were loaded with crystal violet (CV) and ethinyl estradiol (EE) and the drug release studies were performed in 0.01 M phosphate buffered saline (PBS) (pH 7.1-7.4) at 37.0 +/- 1.0 degrees C for upto 4 weeks. The large tubes were loaded with CV and the drug release was studied in vitro in PBS and also ex vivo in rabbit's vitreous humor. Linear release rates with R(2) > 0.9900 were obtained for all groups with CV and EE. Release rates of 7.8 +/- 2.5, 16.2 +/- 5.5, and 22.5 +/- 6.0 ng/day for CV and 30.1 +/- 5.8 ng/day for EE were obtained for small tubes. For large tubes, a release rate of 10.8 +/- 4.1, 15.8 +/- 4.8 and 22.1 +/- 6.7 microg/day was observed in vitro in PBS and a release rate of 5.8 +/- 1.8 microg/day was observed ex vivo in vitreous humor.

Controlled formation and resistivity scaling of nickel silicide nanolines.

We demonstrate a top-down method for fabricating nickel mono-silicide (NiSi) nanolines (also referred to as nanowires) with smooth sidewalls and line widths down to 15 nm. Four-probe electrical measurements reveal that the room temperature electrical resistivity of the NiSi nanolines remains constant as the line widths are reduced to 23 nm. The resistivity at cryogenic temperatures is found to increase with decreasing line width. This finding can be attributed to electron scattering at the sidewalls and is used to deduce an electron mean free path of 6.3 nm for NiSi at room temperature. The results suggest that NiSi nanolines with smooth sidewalls are able to meet the requirements for implementation at the 22 nm technology node without degradation of device performance.

Bilinear Factor Matrix Norm Minimization for Robust PCA: Algorithms and Applications.

The heavy-tailed distributions of corrupted outliers and singular values of all channels in low-level vision have proven effective priors for many applications such as background modeling, photometric stereo and image alignment. And they can be well modeled by a hyper-Laplacian. However, the use of such distributions generally leads to challenging non-convex, non-smooth and non-Lipschitz problems, and makes existing algorithms very slow for large-scale applications. Together with the analytic solutions to $\ell _{p}$ -norm minimization with two specific values of $p$ , i.e., $p=1/2$ and $p=2/3$ , we propose two novel bilinear factor matrix norm minimization models for robust principal component analysis. We first define the double nuclear norm and Frobenius/nuclear hybrid norm penalties, and then prove that they are in essence the Schatten- $1/2$ and $2/3$ quasi-norms, respectively, which lead to much more tractable and scalable Lipschitz optimization problems. Our experimental analysis shows that both our methods yield more accurate solutions than original Schatten quasi-norm minimization, even when the number of observations is very limited. Finally, we apply our penalties to various low-level vision problems, e.g., text removal, moving object detection, image alignment and inpainting, and show that our methods usually outperform the state-of-the-art methods.

Online clustering algorithms for radar emitter classification.

Radar emitter classification is a special application of data clustering for classifying unknown radar emitters from received radar pulse samples. The main challenges of this task are the high dimensionality of radar pulse samples, small sample group size, and closely located radar pulse clusters. In this paper, two new online clustering algorithms are developed for radar emitter classification: One is model-based using the Minimum Description Length (MDL) criterion and the other is based on competitive learning. Computational complexity is analyzed for each algorithm and then compared. Simulation results show the superior performance of the model-based algorithm over competitive learning in terms of better classification accuracy, flexibility, and stability.

On the Convergence of the LMS Algorithm with Adaptive Learning Rate for Linear Feedforward Networks.

We consider the problem of training a linear feedforward neural network by using a gradient descent-like LMS learning algorithm. The objective is to find a weight matrix for the network, by repeatedly presenting to it a finite set of examples, so that the sum of the squares of the errors is minimized. Kohonen showed that with a small but fixed learning rate (or stepsize) some subsequences of the weight matrices generated by the algorithm will converge to certain matrices close to the optimal weight matrix. In this paper, we show that, by dynamically decreasing the learning rate during each training cycle, the sequence of matrices generated by the algorithm will converge to the optimal weight matrix. We also show that for any given ∊ > 0 the LMS algorithm, with decreasing learning rates, will generate an ∊-optimal weight matrix (i.e., a matrix of distance at most ∊ away from the optimal matrix) after O(1/∊) training cycles. This is in contrast to Ω(1/∊log 1/∊) training cycles needed to generate an ∊-optimal weight matrix when the learning rate is kept fixed. We also give a general condition for the learning rates under which the LMS learning algorithm is guaranteed to converge to the optimal weight matrix.

Nano-opto-mechanical characterization of neuron membrane mechanics under cellular growth and differentiation.

We designed and fabricated silicon probe with nanophotonic force sensor to directly stimulate neurons (PC12) and measured its effect on neurite initiation and elongation. A single-layer pitch-variable diffractive nanogratings was fabricated on silicon nitride probe using e-beam lithography, reactive ion etching and wet-etching techniques. The nanogratings consist of flexure folding beams suspended between two parallel cantilevers of known stiffness. The probe displacement, therefore the force, can be measured through grating transmission spectrum. We measured the mechanical membrane characteristics of PC12 cells using the force sensors with displacement range of 10 mum and force sensitivity 8 muN/mum. Young's moduli of 425 +/- 30 Pa are measured with membrane deflection of 1% for PC12 cells cultured on polydimethylsiloxane (PDMS) substrate coated with collagen or laminin in Ham's F-12K medium. In a series of measurements, we have also observed stimulation of directed neurite contraction up to 6 mum on extended probing for a time period of 30 min. This method is applicable to measure central neurons mechanics under subtle tensions for studies on development and morphogenesis. The close synergy between the nano-photonic measurements and neurological verification can improve our understanding of the effect of external conditions on the mechanical properties of cells during growth and differentiation.