University of Münster

About: University of Münster is a education organization based out in Münster, Germany. It is known for research contribution in the topics: Population & Catalysis. The organization has 35609 authors who have published 69059 publications receiving 2278534 citations. The organization is also known as: University of Munster & University of Muenster.

Attentional Feature Fusion

Abstract: Feature fusion, the combination of features from different layers or branches, is an omnipresent part of modern network architectures. It is often implemented via simple operations, such as summation or concatenation, but this might not be the best choice. In this work, we propose a uniform and general scheme, namely attentional feature fusion, which is applicable for most common scenarios, including feature fusion induced by short and long skip connections as well as within Inception layers. To better fuse features of inconsistent semantics and scales, we propose a multiscale channel attention module, which addresses issues that arise when fusing features given at different scales. We also demonstrate that the initial integration of feature maps can become a bottleneck and that this issue can be alleviated by adding another level of attention, which we refer to as iterative attentional feature fusion. With fewer layers or parameters, our models outperform state-of-the-art networks on both CIFAR-100 and ImageNet datasets, which suggests that more sophisticated attention mechanisms for feature fusion hold great potential to consistently yield better results compared to their direct counterparts. Our codes and trained models are available online1.

Letermovir for cytomegalovirus prophylaxis in hematopoietic-cell transplantation

Abstract: BackgroundCytomegalovirus (CMV) infection is a leading cause of illness and death in patients who have undergone allogeneic hematopoietic-cell transplantation. Available treatments are restricted by clinically significant toxic effects and drug resistance. MethodsIn this phase 2 study, we evaluated the effect of letermovir (also known as AIC246), a new anti-CMV drug with a novel mechanism of action, on the incidence and time to onset of prophylaxis failure in CMV-seropositive recipients of allogeneic hematopoietic-cell transplants from matched related or unrelated donors. From March 2010 through October 2011, we randomly assigned 131 transplant recipients in a 3:1 ratio to three sequential study cohorts according to a double-blind design. Patients received oral letermovir (at a dose of 60, 120, or 240 mg per day, or matching placebo) for 12 weeks after engraftment. The primary end point was all-cause prophylaxis failure, defined as discontinuation of the study drug because of CMV antigen or DNA detection,...

Blood vessel formation and function in bone

Abstract: In addition to their conventional role as a conduit system for gases, nutrients, waste products or cells, blood vessels in the skeletal system play active roles in controlling multiple aspects of bone formation and provide niches for hematopoietic stem cells that reside within the bone marrow. In addition, recent studies have highlighted roles for blood vessels during bone healing. Here, we provide an overview of the architecture of the bone vasculature and discuss how blood vessels form within bone, how their formation is modulated, and how they function during development and fracture repair.

Effect of preoperative chemoradiation in addition to preoperative chemotherapy: a randomised trial in stage III non-small-cell lung cancer

Abstract: Summary Background Preoperative chemotherapy improves survival in patients with stage III non-small-cell lung cancer (NSCLC) amenable to resection. We aimed to assess the additional effect of preoperative chemoradiation on tumour resection, pathological response, and survival in these patients. Methods Between Oct 1, 1995, and July 1, 2003, patients with stage IIIA–IIIB NSCLC and invasive mediastinal assessment from 26 participating institutions of the German Lung Cancer Cooperative Group (GLCCG) were randomly assigned to one of two treatment groups. The intervention group were scheduled to receive three cycles of cisplatin and etoposide, followed by twice-daily radiation with concurrent carboplatin and vindesine, and then surgical resection (those with positive resection margins or unresectable disease were offered further twice-daily radiotherapy). The control group were scheduled to receive three cycles of cisplatin and etoposide, followed by surgery, and then further radiotherapy. The primary endpoint was median progression-free survival (PFS) in patients eligible for treatment after randomisation. Secondary endpoints in patients eligible for treatment after randomisation were overall survival (OS) and the proportion of patients undergoing surgery. Secondary endpoints in patients with tumour resection were the proportion with negative resection margins, the proportion with complete resection, the proportion with histopathological response, and the proportion with mediastinal downstaging. Additionally, exploratory (not prespecified) post-hoc analyses in terms of PFS and OS were done on patients not amenable to resection and on further subgroups of patients undergoing resection. Analyses were by intention to treat. This trial is registered on the ClinicalTrials.gov website, number NCT 00176137. Findings 558 patients were randomly assigned. 34 patients did not meet inclusion criteria and were excluded. Of 524 eligible patients, 142 of 264 (54%) in the interventional group and 154 of 260 (59%) in the control group underwent surgery; 98 of 264 (37%) and 84 of 260 (32%) underwent complete resection. In patients with complete resection, the proportion of those with mediastinal downstaging (45 of 98 [46%] and 24 of 84 [29%], p=0·02) and pathological response (59 of 98 [60%] and 17 of 84 [20%], p vs 10·0 months, range 1·0–111·0 [8·9–11·5], 5-year PFS 16% [11–21] vs 14% [10–19], hazard ratio (HR) 0·99 [0·81–1·19], p=0·87), in those undergoing tumour resection, or in patients with complete resection. In both groups, 35% of patients undergoing surgery received a pneumonectomy (50/142 vs 54/154). In patients receiving a pneumonectomy, treatment-related mortality increased in the interventional group compared with the control group (7/50 [14%] vs 3/54 [6%]). Interpretation In patients with stage III NSCLC amenable to surgery, preoperative chemoradiation in addition to chemotherapy increases pathological response and mediastinal downstaging, but does not improve survival. After induction with chemoradiation, pneumonectomy should be avoided. Funding German Cancer Aid (Bonn, Germany).