Wavelet Approximation-Aware Residual Network for Single Image Deraining.

It has been made great progress on single image deraining based on deep convolutional neural networks (CNNs). In most existing deep deraining methods, CNNs aim to learn a direct mapping from rainy images to clean rain-less images, and their architectures are becoming more and more complex. However, due to the limitation of mixing rain with object edges and background, it is difficult to separate rain and object/background, and the edge details of the image cannot be effectively recovered in the reconstruction process. To address this problem, we propose a novel wavelet approximation-aware residual network (WAAR), wherein rain is effectively removed from both low-frequency structures and high-frequency details at each level separately, especially in low-frequency sub-images at each level. After wavelet transform, we propose novel approximation aware (AAM) and approximation level blending (ALB) mechanisms to further aid the low-frequency networks at each level recover the structure and texture of low-frequency sub-images recursively, while the high frequency network can effectively eliminate rain streaks through block connection and achieve different degrees of edge detail enhancement by adjusting hyperparameters. In addition, we also introduce block connection to enrich the high-frequency details in the high-frequency network, which is favorable for obtaining potential interdependencies between high- and low-frequency features. Experimental results indicate that the proposed WAAR exhibits strong performance in reconstructing clean and rain-free images, recovering real and undistorted texture structures, and enhancing image edges in comparison with the state-of-the-art approaches on synthetic and real image datasets. It shows the effectiveness of our method, especially on image edges and texture details.

Increased expression of SKP2 and phospho-MAPK/ERK1/2 and decreased expression of p27 during tumor progression of cervical neoplasms.

OBJECTIVE: The objective of this study was to investigate whether the expression of SKP2, p27 and phospho-MAPK/ERK1/2 is associated with the progression of human cervical neoplasia. METHODS: We performed immunohistochemical detection to stain formalin-fixed paraffin-embedded cervical tissues with anti-SKP2 and anti-p27 monoclonal antibodies and anti-phospho-p42/44 MAPK antibody. The study sample included 23 normal cervical epithelium, 25 low-grade squamous intraepithelial lesion (LSIL), 19 high-grade squamous intraepithelial lesion (HSIL), and 31 squamous cell carcinomas (SCC). In addition, 14 frozen cervical biopsies, including 1 normal, 6 HSIL, 2 adenocarcinoma and 5 SCC, and a human cervical cancer cell line (HeLa), were analyzed the expression levels of mRNA and protein of SKP2 and p27 by RT-PCR and Western blot analysis, respectively. RESULTS: The expression of SKP2, p27 and phospho-MAPK/ERK1/2 were strongly associated with cervical neoplastic progression (P<0.0001, P=0.006, P=0.003, respectively; Fisher's Exact Test). In addition, SKP2 expression was positively correlated with phospho-MAPK/ERK1/2 expression (Spearman correlation coefficient=0.480, P=0.0002). The association between SKP2 and phospho-MAPK/ERK1/2 was significant after controlling for the four histologic grades (P=0.038, Mantel-Haenszel test). CONCLUSIONS: These results suggest that expression levels of SKP2, p27 and phospho-MAPK/ERK1/2 may serve as markers for progression in human cervical carcinoma and may also play roles in cervical carcinoma progression and cervical carcinogenesis.

Combined cytologic and histologic diagnosis of liver tumors via one-shot aspiration.

BACKGROUND/AIMS: Ultrasound-guided fine needle aspiration is considered to be a safe alternative in diagnosing liver tumors. Combined cytologic and histologic preparation via one-shot fine needle aspiration was evaluated, especially in outpatients. METHODOLOGY: During a 1-year period, 95 patients (including 57 outpatients and 38 inpatients) who had received 97 sessions of fine needle aspiration and had final diagnosis were enrolled. Using a 22-gauge needle, the aspirate from liver tumor was processed for cytology and cell block histology simultaneously in patients without ascites and/or bleeding tendency. RESULTS: From 94 malignant and 3 benign tumors, the accuracy of cytology was 88.7%; while, the difference between outpatients and inpatients was not significant (91.2% vs. 85%). No complication after fine needle aspiration was elicited in both groups. The rates of inadequate specimen preparation for cytology and histology were 0% and 13.4%, the false-negative rates of malignancy were 11.7% for cytology and 16% for histology. The inadequate and false-negative rates of histology were not associated with gender, underlying liver cirrhosis, or tumor size, type, and location. CONCLUSIONS: Ultrasound-guided 22-gauge one-shot fine needle aspiration is safe, convenient and reliable in diagnosing liver tumor, and can be performed in outpatients meticulously.