This paper focuses on structured-output learning using deep neural networks for 3D human pose estimation from monocular images. Our network takes an image and 3D pose as inputs and outputs a score value, which is high when the image-pose pair matches and low otherwise. The network structure consists of a convolutional neural network for image feature extraction, followed by two sub-networks for transforming the image features and pose into a joint embedding. The score function is then the dot-product between the image and pose embeddings. The image-pose embedding and score function are jointly trained using a maximum-margin cost function. Our proposed framework can be interpreted as a special form of structured support vector machines where the joint feature space is discriminatively learned using deep neural networks. We test our framework on the Human3.6m dataset and obtain state-ofthe-art results compared to other recent methods. Finally, we present visualizations of the image-pose embedding space, demonstrating the network has learned a high-level embedding of body-orientation and pose-configuration.
Langerhans cell histiocytosis (LCH) is a neoplasm of myeloid origin characterized by a clonal proliferation of CD1a+/CD207+ dendritic cells. Recurrent BRAF V600E mutation has been reported in LCH. In the present report, we confirm the feasibility of the high-specificity monoclonal antibody VE1 for detecting BRAF V600E mutation in 36/97 (37.1%) retrospectively enrolled patients with LCH; concordant immunohistochemistry and Sanger sequencing results were seen in 94.8% of cases. We then assessed the tumor immune microenvironment status in LCH, and found that the GATA binding protein 3 (GATA3)+/T-bet+ ratio could distinguish between clinical multi-system/single-system (SS) multifocal and SS unifocal LCH. Notably, we found that BRAF V600E mutation is significantly correlated with increased programmed cell death 1 ligand 1 (PDL1) expression and forkhead box protein 3 (FOXP3)+ regulatory T cells (p < 0.001, 0.009, respectively). Moreover, Cox multivariate survival analysis showed that BRAF V600E mutation and PDL1 were independent prognostic factors of poor disease-free survival (DFS) in LCH (hazard ratio [HR] = 2.38, 95% confidence interval [CI] 1.02–5.56, p = 0.044; HR = 3.06, 95%CI 1.14–7.14, p = 0.025, respectively), and the superiority of PDL1 in sensitivity and specificity as biomarker for DFS in LCH was demonstrated by receiver operator characteristic (ROC) curves when compared with BRAF V600E and risk category. Collectively, this study identifies for the first time relationship between BRAF V600E mutation and a suppressive tumor immune microenvironment in LCH, resulting in disruption of host–tumor immune surveillance, which is DFS. Our findings may provide a rationale for combining immunotherapy and BRAF-targeted therapy for treating patients with BRAF V600E mutant LCH.
Langerhans cell histiocytosis (LCH) is a proliferative disease of CD1a /CD207 dendritic cells. Recurrent BRAFV600E and MAP2K1 mutations have been reported in LCH. To investigate the relationship among the mutation, clinical findings, and differentiation status of LCH, respectively, we studied 97 cases of LCH by using Sanger sequencing and immunohistochemistry. The mutually exclusive BRAFV600E and MAP2K1 mutation rates were 32% and 17.5%, respectively. All MAP2K1 mutations were missense mutations without any in-frame deletions; 2 new recurrent missense mutations (ie, p.E38K and p.P105S) were also found. More BRAFV600E and MAP2K1 mutations occurred in children compared with those in adult patients (P = .001), and BRAF mutation was correlated with relapse (P = .009). To the differentiation-related markers, the BRAF/MAP2K1-mut LCH expressed CD14 but rarely expressed CD83 or CD86 (P < .001). On the contrary, BRAF/MAP2K1-wt LCH cells rarely expressed CD14 but expressed CD86, and some also expressed CD83 (P < .001). This indicated that the BRAF/MAP2K1-mut LCH cells had a more immature state than BRAF/MAP2K1-wt LCH cells. Moreover, we also found the BRAFV600E and MAP2K1 mutations were significantly associated with pERK expression (P < .001). Therefore, the RAS/RAF/MEK/ERK pathway might play a more important role in children than in adult patients with LCH.
We propose a joint foreground-background mixture model (FBM) that simultaneously performs background estimation and motion segmentation in complex dynamic scenes. Our FBM consist of a set of location-specific dynamic texture (DT) components, for modeling local background motion, and set of global DT components, for modeling consistent foreground motion. We derive an EM algorithm for estimating the parameters of the FBM. We also apply spatial constraints to the FBM using an Markov random field grid, and derive a corresponding variational approximation for inference. Unlike existing approaches to background subtraction, our FBM does not require a manually selected threshold or a separate training video. Unlike existing motion segmentation techniques, our FBM can segment foreground motions over complex background with mixed motions, and detect stopped objects. Since most dynamic scene datasets only contain videos with a single foreground object over a simple background, we develop a new challenging dataset with multiple foreground objects over complex dynamic backgrounds. In experiments, we show that jointly modeling the background and foreground segments with FBM yields significant improvements in accuracy on both background estimation and motion segmentation, compared to state-of-the-art methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.