Shanghai Jiao Tong University

About: Shanghai Jiao Tong University is a education organization based out in Shanghai, Shanghai, China. It is known for research contribution in the topics: Population & Cancer. The organization has 157524 authors who have published 184620 publications receiving 3451038 citations. The organization is also known as: Shanghai Communications University & Shanghai Jiaotong University.

Regorafenib plus best supportive care versus placebo plus best supportive care in Asian patients with previously treated metastatic colorectal cancer (CONCUR): a randomised, double-blind, placebo-controlled, phase 3 trial

Abstract: Summary Background In the international randomised phase 3 CORRECT trial (NCT01103323), regorafenib significantly improved overall survival versus placebo in patients with treatment-refractory metastatic colorectal cancer. Of the 760 patients in CORRECT, 111 were Asian (mostly Japanese). This phase 3 trial was done to assess regorafenib in a broader population of Asian patients with refractory metastatic colorectal cancer than was studied in CORRECT. Methods In this randomised, double-blind, placebo-controlled, parallel-group, phase 3 trial done in 25 hospitals in mainland China, Hong Kong, South Korea, Taiwan, and Vietnam, we recruited Asian patients aged 18 years or older with progressive metastatic colorectal cancer who had received at least two previous treatment lines or were unable to tolerate standard treatments. Patients had to have an Eastern Cooperative Oncology Group performance status of 0 or 1, life expectancy of at least 3 months, and adequate bone marrow, liver, and renal function, without other uncontrolled medical disorders. We randomly allocated patients (2:1; with a computer-generated unicentric randomisation list [prepared by the study funder] and interactive voice response system; block size of six; stratified by metastatic site [single vs multiple organs] and time from diagnosis of metastatic disease [ vs ≥18 months]) to receive oral regorafenib 160 mg once daily or placebo on days 1–21 of each 28 day cycle; patients in both groups were also to receive best supportive care. Participants, investigators, and the study funder were masked to treatment assignment. The primary endpoint was overall survival, and we analysed data on an intention-to-treat basis. This trial is registered with ClinicalTrials.gov, number NCT01584830. Findings Between April 29, 2012, and Feb 6, 2013, we screened 243 patients and randomly assigned 204 patients to receive either regorafenib (136 [67%]) or placebo (68 [33%]). After a median follow-up of 7·4 months (IQR 4·3–12·2), overall survival was significantly better with regorafenib than it was with placebo (hazard ratio 0·55, 95% CI 0·40–0·77, one-sided p=0·00016; median overall survival 8·8 months [95% CI 7·3–9·8] in the regorafenib group vs 6·3 months [4·8–7·6] in the placebo group). Drug-related adverse events occurred in 132 (97%) of 136 regorafenib recipients and 31 (46%) of 68 placebo recipients. The most frequent grade 3 or higher regorafenib-related adverse events were hand–foot skin reaction (22 [16%] of 136 patients in the regorafenib group vs none in the placebo group), hypertension (15 [11%] vs two [3%] of 68 patients in the placebo group), hyperbilirubinaemia (nine [7%] vs one [1%]), hypophosphataemia (nine [7%] vs none), alanine aminotransferase concentration increases (nine [7%] vs none), aspartate aminotransferase concentration increases (eight [6%] vs none), lipase concentration increases (six [4%] vs one [1%]), and maculopapular rash (six [4%] vs none). Drug-related serious adverse events occurred in 12 (9%) patients in the regorafenib group and three (4%) in the placebo group. Interpretation This phase 3 trial is the second to show an overall survival benefit with regorafenib compared with placebo in patients with treatment-refractory metastatic colorectal cancer, substantiating the role of regorafenib as an important treatment option for patients whose disease has progressed after standard treatments. In this trial, preceding standard treatments did not necessarily include targeted treatments. Adverse events were generally consistent with the known safety profile of regorafenib in this setting. Funding Bayer HealthCare Pharmaceuticals.

Observation of a topological crystalline insulator phase and topological phase transition in Pb 1−x Sn x Te

Abstract: A topological insulator protected by time-reversal symmetry is realized via spin-orbit interaction-driven band inversion. The topological phase in the Bi(1-x)Sb(x) system is due to an odd number of band inversions. A related spin-orbit system, the Pb(1-x)Sn(x)Te, has long been known to contain an even number of inversions based on band theory. Here we experimentally investigate the possibility of a mirror symmetry-protected topological crystalline insulator phase in the Pb(1-x)Sn(x)Te class of materials that has been theoretically predicted to exist in its end compound SnTe. Our experimental results show that at a finite Pb composition above the topological inversion phase transition, the surface exhibits even number of spin-polarized Dirac cone states revealing mirror-protected topological order distinct from that observed in Bi(1-x)Sb(x). Our observation of the spin-polarized Dirac surface states in the inverted Pb(1-x)Sn(x)Te and their absence in the non-inverted compounds related via a topological phase transition provide the experimental groundwork for opening the research on novel topological order in quantum devices.

Symmetric Cross Entropy for Robust Learning With Noisy Labels

Abstract: Training accurate deep neural networks (DNNs) in the presence of noisy labels is an important and challenging task. Though a number of approaches have been proposed for learning with noisy labels, many open issues remain. In this paper, we show that DNN learning with Cross Entropy (CE) exhibits overfitting to noisy labels on some classes (``easy" classes), but more surprisingly, it also suffers from significant under learning on some other classes (``hard" classes). Intuitively, CE requires an extra term to facilitate learning of hard classes, and more importantly, this term should be noise tolerant, so as to avoid overfitting to noisy labels. Inspired by the symmetric KL-divergence, we propose the approach of Symmetric cross entropy Learning (SL), boosting CE symmetrically with a noise robust counterpart Reverse Cross Entropy (RCE). Our proposed SL approach simultaneously addresses both the under learning and overfitting problem of CE in the presence of noisy labels. We provide a theoretical analysis of SL and also empirically show, on a range of benchmark and real-world datasets, that SL outperforms state-of-the-art methods. We also show that SL can be easily incorporated into existing methods in order to further enhance their performance.

An Overview of Sustainable Green 5G Networks

Abstract: The stringent requirements of a 1000x increase in data traffic and 1 ms round-trip latency have made limiting the potentially tremendous ensuing energy consumption one of the most challenging problems for the design of the upcoming 5G networks. To enable sustainable 5G networks, new technologies have been proposed to improve the system energy efficiency, and alternative energy sources are introduced to reduce our dependence on traditional fossil fuels. In particular, various 5G techniques target the reduction of the energy consumption without sacrificing the quality of service. Meanwhile, energy harvesting technologies, which enable communication transceivers to harvest energy from various renewable resources and ambient radio frequency signals for communication, have drawn significant interest from both academia and industry. In this article, we provide an overview of the latest research on both green 5G techniques and energy harvesting for communication. In addition, some technical challenges and potential research topics for realizing sustainable green 5G networks are also identified.