Search

Heterogeneous graph embedding aims at learning low-dimensional representations from a graph featuring nodes and edges of diverse natures, and meanwhile preserving the underlying topology. Existing approaches along this line have largely relied on meta-paths, which are by nature hand-crafted and pre-defined transition rules, so as to explore the semantics of a graph. Despite the promising results, defining meta-paths requires domain knowledge, and thus when the test distribution deviates from the priors, such methods are prone to errors. In this paper, we propose a self-learning scheme for heterogeneous graph embedding, termed as self-guided walk (SILK), that bypasses meta-paths and learns adaptive attentions for node walking. SILK assumes no prior knowledge or annotation is provided, and conducts a customized random walk to encode the contexts of the heterogeneous graph of interest. Specifically, this is achieved via maintaining a dynamically-updated guidance matrix that records the node-conditioned transition potentials. Experimental results on four real-world datasets demonstrate that SILK significantly outperforms state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Video highlight detection (VHD) has emerged as a pressing task due to the unprecedentedly increasing amount of video data, such as those from e-commerce live-broadcasting platforms. Many approaches focus on exploiting text data, in the form of video description or time-sync comments, to facilitate the VHD task. Despite the promising results, they have largely overlooked the noises inherent in the text data and have mostly relied on isolating the feature of text and video. In this paper, we introduce a novel model to handle VHD, termed Coevolution Network (CoEvo-Net), that allows us to account for joint learning of the language and video features explicitly via a coevolution paradigm, in which features from the two data modalities progressively refine each other. This is achieved by a dedicated CoEvo-Cell that takes language and video together as inputs, extracts cross-modality, and filters the undesired parts of the input, such as words in a sentence. Furthermore, we release a large-scale dataset of e-commerce for VHD, in which each video is coupled with a sentence for description, to benchmark the sentence-based VHD approaches. Extensive experiments on the released dataset demonstrate that CoEvo-Net achieves state-of-the-art performance. Our dataset and code will be made publicly available. [ABSTRACT FROM AUTHOR]

Aims: To compare the fibroinflammatory diseases Kimura's disease (KD) and immunoglobulin (Ig)G4‐related disease (IgG4RD) and to explore their possible relationship. Methods and results: Forty‐six cases of KD and 29 IgG4RD from our institution diagnosed from 2011 to 2020 were studied. They were compared with each other on clinical, pathological and immunohistological features. There were similar clinical features, except that IgG4RD affected an older patient population, with more frequent salivary gland involvement and KD affected head and neck lymph nodes, and showed blood eosinophilia more frequently than IgG4RD. IgG4RD exhibited frequent storiform fibrosis and obliterative phlebitis, while KD showed more frequent tissue eosinophilia, eosinophilic abscess, germinal centre eosinophilic deposit and vascularisation. Twenty to 30% of KD had more than 50 IgG4+ plasma cells (PC) per high‐power field (HPF) and IgG4/IgG+PC ratio exceeding 40%. These parameters, however, occurred in 100% of IgG4RD. Significantly more KD had >10 IgE+PC/HPF and lymphoid germinal centre IgE reticular staining compared to IgG4RD. All these histological and immunohistological features are overlapping in the two diseases, although they differed with statistical significance. Conclusion: Our study confirmed that there is significant overlap in clinical, pathological and immunohistological features between KD and IgG4RD. It is important to recognise these overlapping features, and correlation with a clinicopathological picture is required in differential diagnoses. The overlapping features also suggest a possible close relationship between KD and IgG4RD, which could represent different facets of a continuous fibroinflammatory disease spectrum. [ABSTRACT FROM AUTHOR]

Background: Malignant gastrointestinal neuroectodermal tumor (GNET) is a rare malignant mesenchymal neoplasm that commonly arises in the small bowel, stomach or colon. Meanwhile, SMARCA4-deficient undifferentiated carcinoma is a rarely reported entity with highly aggressive behavior that may involve the ovary, lung, gastrointestinal (GI) tract, endometrium and other organs. To our knowledge, we describe for the first time, an extremely rare case of synchronous GNET and SMARCA4-deficient undifferentiated carcinoma with independent origins in the small intestine. Case Presentation: A 46-year-old woman presented with multiple small intestine masses and underwent surgical resection. Two distinct entities, GNET and SMARCA4-deficient undifferentiated carcinoma, were identified. GNET was composed of epithelioid and spindle cells with clear or eosinophilic cytoplasm arranged in sheets, nest, papillary, fascicular, palisade, rosette like or pseudoalveolar pattern. The neoplastic cells were positive for S-100 and SOX-10. Ewing sarcoma breakpoint region 1 gene (EWSR1) rearrangement was confirmed by fluorescence in situ hybridization (FISH), and EWSR1-CREB1 fusion was revealed by next-generation sequencing (NGS). SMARCA4-deficient undifferentiated carcinoma was composed mainly of poorly adhesive rhabdoid cells with eosinophilic cytoplasm arranged in a diffuse pattern. Multifocal necrosis, brisk mitotic figures as well as multinucleated tumor cells were observed. The neoplastic cells diffusely expressed pancytokeratin and vimentin, and was negative for SMARCA4(BRG1). Frame shift mutation of SMARCA4 was detected by NGS. Conclusions: This is the first report that GNET and SMARCA4-deficient undifferentiated carcinoma occurred simultaneously in the small intestine, with the latter showing multiple involvement of the jejunum and ileum. The potential mechanism underlying co-existence of these two rare malignancies is unknown and need further investigations and concern. [ABSTRACT FROM AUTHOR]