Self-Supervised Information Bottleneck for Deep Multi-View Subspace Clustering.

In this paper, we explore the problem of deep multi-view subspace clustering framework from an information-theoretic point of view. We extend the traditional information bottleneck principle to learn common information among different views in a self-supervised manner, and accordingly establish a new framework called Self-supervised Information Bottleneck based Multi-view Subspace Clustering (SIB-MSC). Inheriting the advantages from information bottleneck, SIB-MSC can learn a latent space for each view to capture common information among the latent representations of different views by removing superfluous information from the view itself while retaining sufficient information for the latent representations of other views. Actually, the latent representation of each view provides a kind of self-supervised signal for training the latent representations of other views. Moreover, SIB-MSC attempts to disengage the other latent space for each view to capture the view-specific information by introducing mutual information based regularization terms, so as to further improve the performance of multi-view subspace clustering. Extensive experiments on real-world multi-view data demonstrate that our method achieves superior performance over the related state-of-the-art methods.

Case report: Right ventricular outflow tract obstruction caused by multicomponent mesenchymal tumor.

Right ventricular outflow tract obstruction (RVOTO) is a cause of hemodynamic instability that can lead to right ventricular dysfunction. Cardiac tumors located in the right ventricle or surrounding structures can cause RVOTO. Herein, we present a rare case of a 21-year-old male with palpitations due to RVOTO caused by a cardiac multicomponent mesenchymal tumor. The tumor was localized in the right ventricular outflow tract, resulting in right side heart enlargement, tricuspid regurgitation, and RVOTO. Hence, tumor resection was performed. The patient was in a stable condition and discharged home on the 6th post-operative day. However, histopathological examination of the tumor specimen suggested a three-component mesenchymal tumor containing mucinous components, formed blood vessels, and fibrous tissue, which is like an atypical capillary hemangioma. After seven years of follow-up, the patient had no right heart enlargement, tricuspid regurgitation, and tumor recurrence. We believe surgical treatment is effective, and this case will provide a reference for clinicians to treat and evaluate the prognosis of similar three-component mesenchymal cardiac tumor cases in the future.

Glypican-3 as an emerging molecular target for hepatocellular carcinoma gene therapy.

Glypican-3 (GPC-3), a membrane-associated heparan sulfate proteoglycan, plays a crucial role in cell proliferation and metastasis, particularly in hepatocellular carcinoma (HCC) progression, and perhaps is a valuable target for its gene therapy. However, its mechanism remains to be explored. In the present study, the biological behaviors of HCC cells were investigated by interfering GPC-3 gene transcription. After the cells were transfected with specific GPC-3 short hairpin RNA (shRNA), the inhibition of GPC-3 expression was 75.6 % in MHCC-97H or 73.8 % in Huh7 cells at mRNA level; the rates of proliferation and apoptosis were 53.6 and 60.5 % in MHCC-97H or 54.9 and 54.4 % in Huh7 cells, with the cell cycles arrested in the G1 phase; the incidences of cell migration, metastasis, and invasion inhibition were 80.1, 56.4, and 69.1 % in MHCC-97H or 80.9, 59.6, and 58.3 % in Huh7 cells, respectively. The cell biological behaviors were altered by silencing GPC-3 with down-regulation of ß-catenin, insulin-like growth factor-II and vascular endothelial growth factor, and Gli1 up-regulation. The cell proliferation was significantly inhibited (up to 95.11 %) by shRNA plus anti-cancer drugs, suggesting that GPC-3 gene should be a potential target for promoting hepatoma cell apoptosis and inhibiting metastasis through the Wnt/ß-catenin and Hh singling pathways.

Negative photoconductivity and memory effects of germanium nanocrystals embedded in HfO2 dielectric.

A metal-insulator-semiconductor (MIS) structure containing an HfO2/SiO2 stack tunnel layer, isolated Germanium (Ge) nanocrystals, and an HfO2 capping layer, was obtained by an electron-beam evaporation method. A high-resolution transmission electron microscopy (HRTEM) study revealed that uniform and pronounced Ge nanocrystals had formed after annealing. Raman spectroscopy provided evidence for the formation of Ge-Ge bonds and the optimal annealing temperature for the crystallization ratio of the Ge. The electric properties of the MIS structure were characterized by capacitance-voltage (C-V) and current-voltage (I-V) measurements at room temperature. Negative photoconductivity was observed when the structure was under a forward bias, which screened the bias voltage, resulting in a decrease in the current at a given voltage and a negative shift in flat band voltage. A relatively high stored charge density of 3.27 x 10(12) cm 2 was also achieved.