Showing papers by "The Chinese University of Hong Kong published in 2015"

Deep Learning Face Attributes in the Wild

Abstract: Predicting face attributes in the wild is challenging due to complex face variations. We propose a novel deep learning framework for attribute prediction in the wild. It cascades two CNNs, LNet and ANet, which are fine-tuned jointly with attribute tags, but pre-trained differently. LNet is pre-trained by massive general object categories for face localization, while ANet is pre-trained by massive face identities for attribute prediction. This framework not only outperforms the state-of-the-art with a large margin, but also reveals valuable facts on learning face representation. (1) It shows how the performances of face localization (LNet) and attribute prediction (ANet) can be improved by different pre-training strategies. (2) It reveals that although the filters of LNet are fine-tuned only with image-level attribute tags, their response maps over entire images have strong indication of face locations. This fact enables training LNet for face localization with only image-level annotations, but without face bounding boxes or landmarks, which are required by all attribute recognition works. (3) It also demonstrates that the high-level hidden neurons of ANet automatically discover semantic concepts after pre-training with massive face identities, and such concepts are significantly enriched after fine-tuning with attribute tags. Each attribute can be well explained with a sparse linear combination of these concepts.

3D ShapeNets: A deep representation for volumetric shapes

Abstract: 3D shape is a crucial but heavily underutilized cue in today's computer vision systems, mostly due to the lack of a good generic shape representation. With the recent availability of inexpensive 2.5D depth sensors (e.g. Microsoft Kinect), it is becoming increasingly important to have a powerful 3D shape representation in the loop. Apart from category recognition, recovering full 3D shapes from view-based 2.5D depth maps is also a critical part of visual understanding. To this end, we propose to represent a geometric 3D shape as a probability distribution of binary variables on a 3D voxel grid, using a Convolutional Deep Belief Network. Our model, 3D ShapeNets, learns the distribution of complex 3D shapes across different object categories and arbitrary poses from raw CAD data, and discovers hierarchical compositional part representation automatically. It naturally supports joint object recognition and shape completion from 2.5D depth maps, and it enables active object recognition through view planning. To train our 3D deep learning model, we construct ModelNet - a large-scale 3D CAD model dataset. Extensive experiments show that our 3D deep representation enables significant performance improvement over the-state-of-the-arts in a variety of tasks.

Association analyses identify 38 susceptibility loci for inflammatory bowel disease and highlight shared genetic risk across populations

Abstract: Ulcerative colitis and Crohn's disease are the two main forms of inflammatory bowel disease (IBD). Here we report the first trans-ancestry association study of IBD, with genome-wide or Immunochip genotype data from an extended cohort of 86,640 European individuals and Immunochip data from 9,846 individuals of East Asian, Indian or Iranian descent. We implicate 38 loci in IBD risk for the first time. For the majority of the IBD risk loci, the direction and magnitude of effect are consistent in European and non-European cohorts. Nevertheless, we observe genetic heterogeneity between divergent populations at several established risk loci driven by differences in allele frequency (NOD2) or effect size (TNFSF15 and ATG16L1) or a combination of these factors (IL23R and IRGM). Our results provide biological insights into the pathogenesis of IBD and demonstrate the usefulness of trans-ancestry association studies for mapping loci associated with complex diseases and understanding genetic architecture across diverse populations.

The status, challenges, and future of additive manufacturing in engineering

Abstract: Additive manufacturing (AM) is poised to bring about a revolution in the way products are designed, manufactured, and distributed to end users. This technology has gained significant academic as well as industry interest due to its ability to create complex geometries with customizable material properties. AM has also inspired the development of the maker movement by democratizing design and manufacturing. Due to the rapid proliferation of a wide variety of technologies associated with AM, there is a lack of a comprehensive set of design principles, manufacturing guidelines, and standardization of best practices. These challenges are compounded by the fact that advancements in multiple technologies (for example materials processing, topology optimization) generate a "positive feedback loop" effect in advancing AM. In order to advance research interest and investment in AM technologies, some fundamental questions and trends about the dependencies existing in these avenues need highlighting. The goal of our review paper is to organize this body of knowledge surrounding AM, and present current barriers, findings, and future trends significantly to the researchers. We also discuss fundamental attributes of AM processes, evolution of the AM industry, and the affordances enabled by the emergence of AM in a variety of areas such as geometry processing, material design, and education. We conclude our paper by pointing out future directions such as the "print-it-all" paradigm, that have the potential to re-imagine current research and spawn completely new avenues for exploration. The fundamental attributes and challenges/barriers of Additive Manufacturing (AM).The evolution of research on AM with a focus on engineering capabilities.The affordances enabled by AM such as geometry, material and tools design.The developments in industry, intellectual property, and education-related aspects.The important future trends of AM technologies.

Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study

Abstract: Current reports on acute kidney injury (AKI) in the intensive care unit (ICU) show wide variation in occurrence rate and are limited by study biases such as use of incomplete AKI definition, selected cohorts, or retrospective design. Our aim was to prospectively investigate the occurrence and outcomes of AKI in ICU patients. The Acute Kidney Injury–Epidemiologic Prospective Investigation (AKI-EPI) study was an international cross-sectional study performed in 97 centers on patients during the first week of ICU admission. We measured AKI by Kidney Disease: Improving Global Outcomes (KDIGO) criteria, and outcomes at hospital discharge. A total of 1032 ICU patients out of 1802 [57.3 %; 95 % confidence interval (CI) 55.0–59.6] had AKI. Increasing AKI severity was associated with hospital mortality when adjusted for other variables; odds ratio of stage 1 = 1.679 (95 % CI 0.890–3.169; p = 0.109), stage 2 = 2.945 (95 % CI 1.382–6.276; p = 0.005), and stage 3 = 6.884 (95 % CI 3.876–12.228; p < 0.001). Risk-adjusted rates of AKI and mortality were similar across the world. Patients developing AKI had worse kidney function at hospital discharge with estimated glomerular filtration rate less than 60 mL/min/1.73 m2 in 47.7 % (95 % CI 43.6–51.7) versus 14.8 % (95 % CI 11.9–18.2) in those without AKI, p < 0.001. This is the first multinational cross-sectional study on the epidemiology of AKI in ICU patients using the complete KDIGO criteria. We found that AKI occurred in more than half of ICU patients. Increasing AKI severity was associated with increased mortality, and AKI patients had worse renal function at the time of hospital discharge. Adjusted risks for AKI and mortality were similar across different continents and regions.

Assessment of Liver Function in Patients With Hepatocellular Carcinoma: A New Evidence-Based Approach—The ALBI Grade

Abstract: Purpose Most patients with hepatocellular carcinoma (HCC) have associated chronic liver disease, the severity of which is currently assessed by the Child-Pugh (C-P) grade. In this international collaboration, we identify objective measures of liver function/dysfunction that independently influence survival in patients with HCC and then combine these into a model that could be compared with the conventional C-P grade. Patients and Methods We developed a simple model to assess liver function, based on 1,313 patients with HCC of all stages from Japan, that involved only serum bilirubin and albumin levels. We then tested the model using similar cohorts from other geographical regions (n = 5,097) and other clinical situations (patients undergoing resection [n = 525] or sorafenib treatment for advanced HCC [n = 1,132]). The specificity of the model for liver (dys)function was tested in patients with chronic liver disease but without HCC (n = 501). Results The model, the Albumin-Bilirubin (ALBI) grade, performed...

Combined Measurement of the Higgs Boson Mass in pp Collisions at √s=7 and 8 TeV with the ATLAS and CMS Experiments

Abstract: A measurement of the Higgs boson mass is presented based on the combined data samples of the ATLAS and CMS experiments at the CERN LHC in the H→γγ and H→ZZ→4l decay channels. The results are obtained from a simultaneous fit to the reconstructed invariant mass peaks in the two channels and for the two experiments. The measured masses from the individual channels and the two experiments are found to be consistent among themselves. The combined measured mass of the Higgs boson is mH=125.09±0.21 (stat)±0.11 (syst) GeV.

Afatinib versus cisplatin-based chemotherapy for EGFR mutation-positive lung adenocarcinoma (LUX-Lung 3 and LUX-Lung 6): analysis of overall survival data from two randomised, phase 3 trials

Abstract: Summary Background We aimed to assess the effect of afatinib on overall survival of patients with EGFR mutation-positive lung adenocarcinoma through an analysis of data from two open-label, randomised, phase 3 trials. Methods Previously untreated patients with EGFR mutation-positive stage IIIB or IV lung adenocarcinoma were enrolled in LUX-Lung 3 (n=345) and LUX-Lung 6 (n=364). These patients were randomly assigned in a 2:1 ratio to receive afatinib or chemotherapy (pemetrexed-cisplatin [LUX-Lung 3] or gemcitabine-cisplatin [LUX-Lung 6]), stratified by EGFR mutation (exon 19 deletion [del19], Leu858Arg, or other) and ethnic origin (LUX-Lung 3 only). We planned analyses of mature overall survival data in the intention-to-treat population after 209 (LUX-Lung 3) and 237 (LUX-Lung 6) deaths. These ongoing studies are registered with ClinicalTrials.gov, numbers NCT00949650 and NCT01121393. Findings Median follow-up in LUX-Lung 3 was 41 months (IQR 35–44); 213 (62%) of 345 patients had died. Median follow-up in LUX-Lung 6 was 33 months (IQR 31–37); 246 (68%) of 364 patients had died. In LUX-Lung 3, median overall survival was 28·2 months (95% CI 24·6–33·6) in the afatinib group and 28·2 months (20·7–33·2) in the pemetrexed-cisplatin group (HR 0·88, 95% CI 0·66–1·17, p=0·39). In LUX-Lung 6, median overall survival was 23·1 months (95% CI 20·4–27·3) in the afatinib group and 23·5 months (18·0–25·6) in the gemcitabine-cisplatin group (HR 0·93, 95% CI 0·72–1·22, p=0·61). However, in preplanned analyses, overall survival was significantly longer for patients with del19-positive tumours in the afatinib group than in the chemotherapy group in both trials: in LUX-Lung 3, median overall survival was 33·3 months (95% CI 26·8–41·5) in the afatinib group versus 21·1 months (16·3–30·7) in the chemotherapy group (HR 0·54, 95% CI 0·36–0·79, p=0·0015); in LUX-Lung 6, it was 31·4 months (95% CI 24·2–35·3) versus 18·4 months (14·6–25·6), respectively (HR 0·64, 95% CI 0·44–0·94, p=0·023). By contrast, there were no significant differences by treatment group for patients with EGFR Leu858Arg-positive tumours in either trial: in LUX-Lung 3, median overall survival was 27·6 months (19·8–41·7) in the afatinib group versus 40·3 months (24·3–not estimable) in the chemotherapy group (HR 1·30, 95% CI 0·80–2·11, p=0·29); in LUX-Lung 6, it was 19·6 months (95% CI 17·0–22·1) versus 24·3 months (19·0–27·0), respectively (HR 1·22, 95% CI 0·81–1·83, p=0·34). In both trials, the most common afatinib-related grade 3–4 adverse events were rash or acne (37 [16%] of 229 patients in LUX-Lung 3 and 35 [15%] of 239 patients in LUX-Lung 6), diarrhoea (33 [14%] and 13 [5%]), paronychia (26 [11%] in LUX-Lung 3 only), and stomatitis or mucositis (13 [5%] in LUX-Lung 6 only). In LUX-Lung 3, neutropenia (20 [18%] of 111 patients), fatigue (14 [13%]) and leucopenia (nine [8%]) were the most common chemotherapy-related grade 3–4 adverse events, while in LUX-Lung 6, the most common chemotherapy-related grade 3–4 adverse events were neutropenia (30 [27%] of 113 patients), vomiting (22 [19%]), and leucopenia (17 [15%]). Interpretation Although afatinib did not improve overall survival in the whole population of either trial, overall survival was improved with the drug for patients with del19 EGFR mutations. The absence of an effect in patients with Leu858Arg EGFR mutations suggests that EGFR del19-positive disease might be distinct from Leu858Arg-positive disease and that these subgroups should be analysed separately in future trials. Funding Boehringer Ingelheim.

Cross-scene crowd counting via deep convolutional neural networks

Abstract: Cross-scene crowd counting is a challenging task where no laborious data annotation is required for counting people in new target surveillance crowd scenes unseen in the training set. The performance of most existing crowd counting methods drops significantly when they are applied to an unseen scene. To address this problem, we propose a deep convolutional neural network (CNN) for crowd counting, and it is trained alternatively with two related learning objectives, crowd density and crowd count. This proposed switchable learning approach is able to obtain better local optimum for both objectives. To handle an unseen target crowd scene, we present a data-driven method to finetune the trained CNN model for the target scene. A new dataset including 108 crowd scenes with nearly 200,000 head annotations is introduced to better evaluate the accuracy of cross-scene crowd counting methods. Extensive experiments on the proposed and another two existing datasets demonstrate the effectiveness and reliability of our approach.

Action recognition with trajectory-pooled deep-convolutional descriptors

Abstract: Visual features are of vital importance for human action understanding in videos. This paper presents a new video representation, called trajectory-pooled deep-convolutional descriptor (TDD), which shares the merits of both hand-crafted features [31] and deep-learned features [24]. Specifically, we utilize deep architectures to learn discriminative convolutional feature maps, and conduct trajectory-constrained pooling to aggregate these convolutional features into effective descriptors. To enhance the robustness of TDDs, we design two normalization methods to transform convolutional feature maps, namely spatiotemporal normalization and channel normalization. The advantages of our features come from (i) TDDs are automatically learned and contain high discriminative capacity compared with those hand-crafted features; (ii) TDDs take account of the intrinsic characteristics of temporal dimension and introduce the strategies of trajectory-constrained sampling and pooling for aggregating deep-learned features. We conduct experiments on two challenging datasets: HMD-B51 and UCF101. Experimental results show that TDDs outperform previous hand-crafted features [31] and deep-learned features [24]. Our method also achieves superior performance to the state of the art on these datasets.

Testing general relativity with present and future astrophysical observations

Abstract: One century after its formulation, Einstein's general relativity (GR) has made remarkable predictions and turned out to be compatible with all experimental tests. Most of these tests probe the theory in the weak-field regime, and there are theoretical and experimental reasons to believe that GR should be modified when gravitational fields are strong and spacetime curvature is large. The best astrophysical laboratories to probe strong-field gravity are black holes and neutron stars, whether isolated or in binary systems. We review the motivations to consider extensions of GR. We present a (necessarily incomplete) catalog of modified theories of gravity for which strong-field predictions have been computed and contrasted to Einstein's theory, and we summarize our current understanding of the structure and dynamics of compact objects in these theories. We discuss current bounds on modified gravity from binary pulsar and cosmological observations, and we highlight the potential of future gravitational wave measurements to inform us on the behavior of gravity in the strong-field regime.

Visual Tracking with Fully Convolutional Networks

Abstract: We propose a new approach for general object tracking with fully convolutional neural network. Instead of treating convolutional neural network (CNN) as a black-box feature extractor, we conduct in-depth study on the properties of CNN features offline pre-trained on massive image data and classification task on ImageNet. The discoveries motivate the design of our tracking system. It is found that convolutional layers in different levels characterize the target from different perspectives. A top layer encodes more semantic features and serves as a category detector, while a lower layer carries more discriminative information and can better separate the target from distracters with similar appearance. Both layers are jointly used with a switch mechanism during tracking. It is also found that for a tracking target, only a subset of neurons are relevant. A feature map selection method is developed to remove noisy and irrelevant feature maps, which can reduce computation redundancy and improve tracking accuracy. Extensive evaluation on the widely used tracking benchmark [36] shows that the proposed tacker outperforms the state-of-the-art significantly.

Saliency detection by multi-context deep learning

Abstract: Low-level saliency cues or priors do not produce good enough saliency detection results especially when the salient object presents in a low-contrast background with confusing visual appearance. This issue raises a serious problem for conventional approaches. In this paper, we tackle this problem by proposing a multi-context deep learning framework for salient object detection. We employ deep Convolutional Neural Networks to model saliency of objects in images. Global context and local context are both taken into account, and are jointly modeled in a unified multi-context deep learning framework. To provide a better initialization for training the deep neural networks, we investigate different pre-training strategies, and a task-specific pre-training scheme is designed to make the multi-context modeling suited for saliency detection. Furthermore, recently proposed contemporary deep models in the ImageNet Image Classification Challenge are tested, and their effectiveness in saliency detection are investigated. Our approach is extensively evaluated on five public datasets, and experimental results show significant and consistent improvements over the state-of-the-art methods.

Deeply learned face representations are sparse, selective, and robust

Abstract: This paper designs a high-performance deep convolutional network (DeepID2+) for face recognition. It is learned with the identification-verification supervisory signal. By increasing the dimension of hidden representations and adding supervision to early convolutional layers, DeepID2+ achieves new state-of-the-art on LFW and YouTube Faces benchmarks. Through empirical studies, we have discovered three properties of its deep neural activations critical for the high performance: sparsity, selectiveness and robustness. (1) It is observed that neural activations are moderately sparse. Moderate sparsity maximizes the discriminative power of the deep net as well as the distance between images. It is surprising that DeepID2+ still can achieve high recognition accuracy even after the neural responses are binarized. (2) Its neurons in higher layers are highly selective to identities and identity-related attributes. We can identify different subsets of neurons which are either constantly excited or inhibited when different identities or attributes are present. Although DeepID2+ is not taught to distinguish attributes during training, it has implicitly learned such high-level concepts. (3) It is much more robust to occlusions, although occlusion patterns are not included in the training set.

Sofosbuvir and Velpatasvir for HCV Genotype 1, 2, 4, 5, and 6 Infection

Abstract: BackgroundA simple treatment regimen that is effective in a broad range of patients who are chronically infected with the hepatitis C virus (HCV) remains an unmet medical need. MethodsWe conducted a phase 3, double-blind, placebo-controlled study involving untreated and previously treated patients with chronic HCV genotype 1, 2, 4, 5, or 6 infection, including those with compensated cirrhosis. Patients with HCV genotype 1, 2, 4, or 6 were randomly assigned in a 5:1 ratio to receive the nucleotide polymerase inhibitor sofosbuvir and the NS5A inhibitor velpatasvir in a once-daily, fixed-dose combination tablet or matching placebo for 12 weeks. Because of the low prevalence of genotype 5 in the study regions, patients with genotype 5 did not undergo randomization but were assigned to the sofosbuvir–velpatasvir group. The primary end point was a sustained virologic response at 12 weeks after the end of therapy. ResultsOf the 624 patients who received treatment with sofosbuvir–velpatasvir, 34% had HCV genotype...

Learning from massive noisy labeled data for image classification

Abstract: Large-scale supervised datasets are crucial to train convolutional neural networks (CNNs) for various computer vision problems. However, obtaining a massive amount of well-labeled data is usually very expensive and time consuming. In this paper, we introduce a general framework to train CNNs with only a limited number of clean labels and millions of easily obtained noisy labels. We model the relationships between images, class labels and label noises with a probabilistic graphical model and further integrate it into an end-to-end deep learning system. To demonstrate the effectiveness of our approach, we collect a large-scale real-world clothing classification dataset with both noisy and clean labels. Experiments on this dataset indicate that our approach can better correct the noisy labels and improves the performance of trained CNNs.

Colorectal cancer screening: a global overview of existing programmes

Abstract: Colorectal cancer (CRC) ranks third among the most commonly diagnosed cancers worldwide, with wide geographical variation in incidence and mortality across the world. Despite proof that screening can decrease CRC incidence and mortality, CRC screening is only offered to a small proportion of the target population worldwide. Throughout the world there are widespread differences in CRC screening implementation status and strategy. Differences can be attributed to geographical variation in CRC incidence, economic resources, healthcare structure and infrastructure to support screening such as the ability to identify the target population at risk and cancer registry availability. This review highlights issues to consider when implementing a CRC screening programme and gives a worldwide overview of CRC burden and the current status of screening programmes, with focus on international differences.

Chitosan: An Update on Potential Biomedical and Pharmaceutical Applications.

Abstract: Chitosan is a natural polycationic linear polysaccharide derived from chitin. The low solubility of chitosan in neutral and alkaline solution limits its application. Nevertheless, chemical modification into composites or hydrogels brings to it new functional properties for different applications. Chitosans are recognized as versatile biomaterials because of their non-toxicity, low allergenicity, biocompatibility and biodegradability. This review presents the recent research, trends and prospects in chitosan. Some special pharmaceutical and biomedical applications are also highlighted.

Compression Artifacts Reduction by a Deep Convolutional Network

Abstract: Lossy compression introduces complex compression artifacts, particularly the blocking artifacts, ringing effects and blurring. Existing algorithms either focus on removing blocking artifacts and produce blurred output, or restores sharpened images that are accompanied with ringing effects. Inspired by the deep convolutional networks (DCN) on super-resolution, we formulate a compact and efficient network for seamless attenuation of different compression artifacts. We also demonstrate that a deeper model can be effectively trained with the features learned in a shallow network. Following a similar "easy to hard" idea, we systematically investigate several practical transfer settings and show the effectiveness of transfer learning in low level vision problems. Our method shows superior performance than the state-of-the-arts both on the benchmark datasets and the real-world use cases (i.e. Twitter).

Clinical activity of afatinib in patients with advanced non-small-cell lung cancer harbouring uncommon EGFR mutations: a combined post-hoc analysis of LUX-Lung 2, LUX-Lung 3, and LUX-Lung 6.

Abstract: Summary Background Most patients with non-small-cell lung cancer tumours that have EGFR mutations have deletion mutations in exon 19 or the Leu858Arg point mutation in exon 21, or both (ie, common mutations). However, a subset of patients (10%) with mutations in EGFR have tumours that harbour uncommon mutations. There is a paucity of data regarding the sensitivity of these tumours to EGFR inhibitors. Here we present data for the activity of afatinib in patients with advanced non-small-cell lung cancer that have tumours harbouring uncommon EGFR mutations. Methods In this post-hoc analysis, we used prospectively collected data from tyrosine kinase inhibitor-naive patients with EGFR mutation-positive advanced (stage IIIb–IV) lung adenocarcinomas who were given afatinib in a single group phase 2 trial (LUX-Lung 2), and randomised phase 3 trials (LUX-Lung 3 and LUX-Lung 6). Analyses were done in the intention-to-treat population, including all randomly assigned patients with uncommon EGFR mutations. The type of EGFR mutation (exon 19 deletion [del19], Leu858Arg point mutation in exon 21, or other) and ethnic origin (LUX-Lung 3 only; Asian vs non-Asian) were pre-specified stratification factors in the randomised trials. We categorised all uncommon mutations as: point mutations or duplications in exons 18–21 (group 1); de-novo Thr790Met mutations in exon 20 alone or in combination with other mutations (group 2); or exon 20 insertions (group 3). We also assessed outcomes in patients with the most frequent uncommon mutations, Gly719Xaa, Leu861Gln, and Ser768Ile, alone or in combination with other mutations. Response was established by independent radiological review. These trials are registered with ClinicalTrials.gov, numbers NCT00525148, NCT00949650, and NCT01121393. Findings Of 600 patients given afatinib across the three trials, 75 (12%) patients had uncommon EGFR mutations (38 in group 1, 14 in group 2, 23 in group 3). 27 (71·1%, 95% CI 54·1–84·6) patients in group 1 had objective responses, as did two (14·3%, 1·8–42·8) in group 2 and two (8·7%, 1·1–28·0) in group 3. Median progression-free survival was 10·7 months (95% CI 5·6–14·7) in group 1, 2·9 months (1·2–8·3) in group 2; and 2·7 months (1·8–4·2) in group 3. Median overall survival was 19·4 months (95% CI 16·4–26·9) in group 1, 14·9 months (8·1–24·9) in group 2, and 9·2 months (4·1–14·2) in group 3. For the most frequent uncommon mutations, 14 (77·8%, 95% CI 52·4–93·6) patients with Gly719Xaa had an objective response, as did nine (56·3%, 29·9–80·2) with Leu861Gln, and eight (100·0%, 63·1–100·0) with Ser768Ile. Interpretation Afatinib was active in non-small-cell lung cancer tumours that harboured certain types of uncommon EGFR mutations, especially Gly719Xaa, Leu861Gln, and Ser768Ile, but less active in other mutations types. Clinical benefit was lower in patients with de-novo Thr790Met and exon 20 insertion mutations. These data could help inform clinical decisions for patients with non-small-cell lung cancer harbouring uncommon EGFR mutations. Funding Boehringer Ingelheim.

Semantic Image Segmentation via Deep Parsing Network

Abstract: This paper addresses semantic image segmentation by incorporating rich information into Markov Random Field (MRF), including high-order relations and mixture of label contexts. Unlike previous works that optimized MRFs using iterative algorithm, we solve MRF by proposing a Convolutional Neural Network (CNN), namely Deep Parsing Network (DPN), which enables deterministic end-to-end computation in a single forward pass. Specifically, DPN extends a contemporary CNN architecture to model unary terms and additional layers are carefully devised to approximate the mean field algorithm (MF) for pairwise terms. It has several appealing properties. First, different from the recent works that combined CNN and MRF, where many iterations of MF were required for each training image during back-propagation, DPN is able to achieve high performance by approximating one iteration of MF. Second, DPN represents various types of pairwise terms, making many existing works as its special cases. Third, DPN makes MF easier to be parallelized and speeded up in Graphical Processing Unit (GPU). DPN is thoroughly evaluated on the PASCAL VOC 2012 dataset, where a single DPN model yields a new state-of-the-art segmentation accuracy of 77.5%.

Middle East Respiratory Syndrome

Abstract: Middle East respiratory syndrome (MERS) is a highly lethal respiratory disease caused by a novel single-stranded, positive-sense RNA betacoronavirus (MERS-CoV). Dromedary camels, hosts for MERS-CoV, are implicated in direct or indirect transmission to human beings, although the exact mode of transmission is unknown. The virus was fi rst isolated from a patient who died from a severe respiratory illness in June, 2012, in Jeddah, Saudi Arabia. As of May 31, 2015, 1180 laboratory-confi rmed cases (483 deaths; 40% mortality) have been reported to WHO. Both community-acquired and hospital-acquired cases have been reported with little human-to-human transmission reported in the community. Although most cases of MERS have occurred in Saudi Arabia and the United Arab Emirates, cases have been reported in Europe, the USA, and Asia in people who travelled from the Middle East or their contacts. Clinical features of MERS range from asymptomatic or mild disease to acute respiratory distress syndrome and multiorgan failure resulting in death, especially in individuals with underlying comorbidities. No specifi c drug treatment exists for MERS and infection prevention and control measures are crucial to prevent spread in health-care facilities. MERS-CoV continues to be an endemic, low-level public health threat. However, the virus could mutate to have increased interhuman transmissibility, increasing its pandemic potential.

A summary of the new GINA strategy: a roadmap to asthma control.

Abstract: Over the past 20 years, the Global Initiative for Asthma (GINA) has regularly published and annually updated a global strategy for asthma management and prevention that has formed the basis for many national guidelines. However, uptake of existing guidelines is poor. A major revision of the GINA report was published in 2014, and updated in 2015, reflecting an evolving understanding of heterogeneous airways disease, a broader evidence base, increasing interest in targeted treatment, and evidence about effective implementation approaches. During development of the report, the clinical utility of recommendations and strategies for their practical implementation were considered in parallel with the scientific evidence.This article provides a summary of key changes in the GINA report, and their rationale. The changes include a revised asthma definition; tools for assessing symptom control and risk factors for adverse outcomes; expanded indications for inhaled corticosteroid therapy; a framework for targeted treatment based on phenotype, modifiable risk factors, patient preference, and practical issues; optimisation of medication effectiveness by addressing inhaler technique and adherence; revised recommendations about written asthma action plans; diagnosis and initial treatment of the asthma-chronic obstructive pulmonary disease overlap syndrome; diagnosis in wheezing pre-school children; and updated strategies for adaptation and implementation of GINA recommendations.

A large-scale car dataset for fine-grained categorization and verification

Abstract: This paper aims to highlight vision related tasks centered around “car”, which has been largely neglected by vision community in comparison to other objects. We show that there are still many interesting car-related problems and applications, which are not yet well explored and researched. To facilitate future car-related research, in this paper we present our on-going effort in collecting a large-scale dataset, “CompCars”, that covers not only different car views, but also their different internal and external parts, and rich attributes. Importantly, the dataset is constructed with a cross-modality nature, containing a surveillance-nature set and a web-nature set. We further demonstrate a few important applications exploiting the dataset, namely car model classification, car model verification, and attribute prediction. We also discuss specific challenges of the car-related problems and other potential applications that worth further investigations. The latest dataset can be downloaded at http://mmlab.ie.cuhk.edu.hk/ datasets/comp_cars/index.html

Chemotherapy and radiotherapy in nasopharyngeal carcinoma: an update of the MAC-NPC meta-analysis

Abstract: Summary Background A previous individual patient data meta-analysis by the Meta-Analysis of Chemotherapy in Nasopharynx Carcinoma (MAC-NPC) collaborative group to assess the addition of chemotherapy to radiotherapy showed that it improves overall survival in nasopharyngeal carcinoma. This benefit was restricted to patients receiving concomitant chemotherapy and radiotherapy. The aim of this study was to update the meta-analysis, include recent trials, and to analyse separately the benefit of concomitant plus adjuvant chemotherapy. Methods We searched PubMed, Web of Science, Cochrane Controlled Trials meta-register, ClinicalTrials.gov, and meeting proceedings to identify published or unpublished randomised trials assessing radiotherapy with or without chemotherapy in patients with non-metastatic nasopharyngeal carcinoma and obtained updated data for previously analysed studies. The primary endpoint of interest was overall survival. All trial results were combined and analysed using a fixed-effects model. The statistical analysis plan was pre-specified in a protocol. All data were analysed on an intention-to-treat basis. Findings We analysed data from 19 trials and 4806 patients. Median follow-up was 7·7 years (IQR 6·2–11·9). We found that the addition of chemotherapy to radiotherapy significantly improved overall survival (hazard ratio [HR] 0·79, 95% CI 0·73–0·86, p Interpretation Our results confirm that the addition of concomitant chemotherapy to radiotherapy significantly improves survival in patients with locoregionally advanced nasopharyngeal carcinoma. To our knowledge, this is the first analysis that examines the effect of concomitant chemotherapy with and without adjuvant chemotherapy as distinct groups. Further studies on the specific benefits of adjuvant chemotherapy after concomitant chemoradiotherapy are needed. Funding French Ministry of Health (Programme d'actions integrees de recherche VADS), Ligue Nationale Contre le Cancer, and Sanofi-Aventis.

Hybrid halide perovskite solar cell precursors: colloidal chemistry and coordination engineering behind device processing for high efficiency.

Abstract: The precursor of solution-processed perovskite thin films is one of the most central components for high-efficiency perovskite solar cells. We first present the crucial colloidal chemistry visualization of the perovskite precursor solution based on analytical spectra and reveal that perovskite precursor solutions for solar cells are generally colloidal dispersions in a mother solution, with a colloidal size up to the mesoscale, rather than real solutions. The colloid is made of a soft coordination complex in the form of a lead polyhalide framework between organic and inorganic components and can be structurally tuned by the coordination degree, thereby primarily determining the basic film coverage and morphology of deposited thin films. By utilizing coordination engineering, particularly through employing additional methylammonium halide over the stoichiometric ratio for tuning the coordination degree and mode in the initial colloidal solution, along with a thermal leaching for the selective release of ex...