内蒙古民族大学

The Zero-Shot Sketch-Based Image Retrieval (ZS-SBIR) task aims to retrieve images associated with sketches from unseen classes, bringing great convenience to the engineering field. To address the modality gap, most existing works project images and sketches into a shared Euclidean space. However, the hierarchical structure of image data makes the Euclidean space not the optimal choice as an embedding space for representing complex structured image data. Meanwhile, existing text and hierarchical models are not effective enough for addressing the problem of knowledge transfer. To address these issues, this article proposes an original Hyperbolic-Based Cross-Modal Semantic Remodeling Network (called HCMSN) for ZS-SBIR. Specifically, this article proposes to extract category-level word embeddings based on BERT model, then align image features and sketches with the word embeddings using adversarial methods. Meanwhile, this article further proposes a cross-modal retrieval feature reconstruction network for improving the informativeness and robustness of retrieval features. Moreover, this article presents a feature projection network that maps the retrieval features to the hyperbolic space to generate the hyperbolic retrieval features, thus effectively representing the data with hierarchical structure. Extensive experiments demonstrate that the mAP@all of our HCMSN model surpasses CNN-based models by 20.9% on the Sketchy dataset, 1.2% on the more difficult TU-Berlin dataset, and 13.6% on the more challenging QuickDraw dataset.

Mesial temporal lobe epilepsy (mTLE), the most common type of temporal lobe epilepsy (TLE), is particularly relevant due to its high frequency of therapeutic resistance of anti-epileptic therapies. MicroRNAs (miRNAs) have been shown to be dysregulated in epilepsy and neurodegenerative diseases, and we hypothesized that miRNAs could be involved in the pathogenesis of MTLE. The present study aimed to explore the expression and functions of miRNA-153 in mTLE. The expression levels of miRNA-153 in refractory TLE patients were evaluated. The bioinformatics analysis showed that the potential target genes of miR-153 were involved in biological processes, molecular functions, and cellular components. miRNA-153 is significantly dysregulated in temporal cortex and plasma of mTLE patients. We identify HIF-1α as a direct target of miRNA-153, and luciferase reporter assays demonstrated that miR-153 could regulate the HIF-1αexpression via 3'-UTR pairing. These data suggest that miR-153 might represent a useful biomarker and treatment target for patients with mTLE.

[This corrects the article DOI: 10.18632/oncotarget.23119.].