High-Similarity-Pass Attention for Single Image Super-Resolution.

Recent developments in the field of non-local attention (NLA) have led to a renewed interest in self-similarity-based single image super-resolution (SISR). Researchers usually use the NLA to explore non-local self-similarity (NSS) in SISR and achieve satisfactory reconstruction results. However, a surprising phenomenon that the reconstruction performance of the standard NLA is similar to that of the NLA with randomly selected regions prompted us to revisit NLA. In this paper, we first analyzed the attention map of the standard NLA from different perspectives and discovered that the resulting probability distribution always has full support for every local feature, which implies a statistical waste of assigning values to irrelevant non-local features, especially for SISR which needs to model long-range dependence with a large number of redundant non-local features. Based on these findings, we introduced a concise yet effective soft thresholding operation to obtain high-similarity-pass attention (HSPA), which is beneficial for generating a more compact and interpretable distribution. Furthermore, we derived some key properties of the soft thresholding operation that enable training our HSPA in an end-to-end manner. The HSPA can be integrated into existing deep SISR models as an efficient general building block. In addition, to demonstrate the effectiveness of the HSPA, we constructed a deep high-similarity-pass attention network (HSPAN) by integrating a few HSPAs in a simple backbone. Extensive experimental results demonstrate that HSPAN outperforms state-of-the-art approaches on both quantitative and qualitative evaluations. Our code and a pre-trained model were uploaded to GitHub (https://github.com/laoyangui/HSPAN) for validation.

Coexisting Infantile Hepatic Hemangioma and Hepatoblastoma in a Neonate: A Case Report.

Infantile hepatic hemangioma and hepatoblastoma are the most common benign and malignant tumors of the liver in the neonatal and early childhood periods, respectively. However, the simultaneous occurrence of these 2 tumors in the same liver lesion is very rare. We report a case of a newborn infant diagnosed with a liver mass by ultrasound examination 4 days after birth. Serum alpha-fetoprotein (AFP) was elevated for his age (32,881.7 ng/mL). The liver mass was resected. Macroscopically, an externally protruding mass measuring 6 × 4 × 3.5 cm was identified. Microscopically, we observed the coexistence of infantile hepatic hemangioma and epithelial hepatoblastoma components within the tumor. The infantile hepatic hemangioma component was composed of multiple small vascular channels lined by endothelial cells. In the hepatoblastoma component, tumor cells were arranged in a 2- to 3-cell-thick trabecular formation. Immunohistochemistry indicated that the tumor cells in the infantile hepatic hemangioma component expressed CD34, CD31, FLI1, and ERG, and those in the hepatoblastoma component expressed hepatocyte, keratin AE1/AE3 and keratin 8, glypican 3, glutamine synthetase, and AFP. Pathological examination confirmed the presence of an infantile hepatic hemangioma combined with epithelial hepatoblastoma (fetal type). The boy did not undergo chemotherapy after the operation. Regular follow-up through serum AFP levels and liver ultrasound for 16 months to date show that the serum AFP levels decreased continuously to normal levels, with no signs of tumor recurrence or metastasis. The coexistence of infantile hepatic hemangioma and hepatoblastoma is rare. Hepatoblastoma should be considered in neonates with liver tumors and elevated AFP.

Global Learnable Attention for Single Image Super-Resolution.

Self-similarity is valuable to the exploration of non-local textures in single image super-resolution (SISR). Researchers usually assume that the importance of non-local textures is positively related to their similarity scores. In this paper, we surprisingly found that when repairing severely damaged query textures, some non-local textures with low-similarity which are closer to the target can provide more accurate and richer details than the high-similarity ones. In these cases, low-similarity does not mean inferior but is usually caused by different scales or orientations. Utilizing this finding, we proposed a Global Learnable Attention (GLA) to adaptively modify similarity scores of non-local textures during training instead of only using a fixed similarity scoring function such as the dot product. The proposed GLA can explore non-local textures with low-similarity but more accurate details to repair severely damaged textures. Furthermore, we propose to adopt Super-Bit Locality-Sensitive Hashing (SB-LSH) as a preprocessing method for our GLA. With the SB-LSH, the computational complexity of our GLA is reduced from quadratic to asymptotic linear with respect to the image size. In addition, the proposed GLA can be integrated into existing deep SISR models as an efficient general building block. Based on the GLA, we constructed a Deep Learnable Similarity Network (DLSN), which achieves state-of-the-art performance for SISR tasks of different degradation types (e.g., blur and noise). Our code and a pre-trained DLSN have been uploaded to GitHub† for validation.

Online Identification of Nonlinear Systems With Separable Structure.

Separable nonlinear models (SNLMs) are of great importance in system modeling, signal processing, and machine learning because of their flexible structure and excellent description of nonlinear behaviors. The online identification of such models is quite challenging, and previous related work usually ignores the special structure where the estimated parameters can be partitioned into a linear and a nonlinear part. In this brief, we propose an efficient first-order recursive algorithm for SNLMs by introducing the variable projection (VP) step. The proposed algorithm utilizes the recursive least-squares method to eliminate the linear parameters, resulting in a reduced function. Then, the stochastic gradient descent (SGD) algorithm is employed to update the parameters of the reduced function. By considering the tight coupling relationship between linear parameters and nonlinear parameters, the proposed first-order VP algorithm is more efficient and robust than the traditional SGD algorithm and alternating optimization algorithm. More importantly, since the proposed algorithm just uses the first-order information, it is easier to apply it to large-scale models. Numerical results on examples of different sizes confirm the effectiveness and efficiency of the proposed algorithm.

Coexistent Kaposi sarcoma and post-transplant lymphoproliferative disorder in the same lymph nodes after pediatric liver transplantation: A case report.

BACKGROUND: Kaposi sarcoma and post-transplant lymphoproliferative disorder have been occasionally reported in post-liver transplant patients. However, the simultaneous occurrence of these two diseases in the same lymph nodes is very rare. CASE SUMMARY: We report the case of a 19-mo-old boy, who presented with intermittent fever and enlarged cervical lymph nodes after liver transplantation. Six cervical lymph nodes were biopsied, and the histopathological examinations revealed multifocal hyperplasia of spindle cells around small blood vessels, extravasated erythrocytes, and heavy infiltration of plasma cells in the cortex and medulla of the lymph nodes. The immunohistochemical analyses of spindle cells revealed positive expression of CD34, CD31, erythroblast transformation-specific-related gene, friend leukemia integration 1, and human herpesvirus-8. The lymphoproliferative lesions expressed CD38, CD138, and multiple myeloma 1. Epstein-Barr encoded RNA in situ hybridization demonstrated Epstein-Barr virus-positive lymphoid cells. Finally, we diagnosed the coexistence of Kaposi sarcoma and post-transplant lymphoproliferative disorder (plasmacytic hyperplasia) in the same lymph nodes. Treatment strategy included anti-CD20 monoclonal antibody (rituximab) and discontinuation of the immunosuppressant therapies. Lymph node biopsies during follow-up examinations revealed lymphoid hyperplasia. CONCLUSION: The rare coexistence of Kaposi sarcoma and post-transplant lymphoproliferative disorder in the same lymph nodes post-liver transplantation possibly associates with immunodeficiency and Epstein-Barr virus and human herpesvirus-8 coinfection.

Bladder paraganglioma after kidney transplantation: A case report.

BACKGROUND: Kidney transplantation is associated with an increased risk of tumors in the urinary bladder. Among all the pathological types of tumors in the bladder, paraganglioma, which arises from extra-adrenal paraganglia and consists of chromaffin cells, is rare. Paragangliomas might cause severe clinical symptoms due to catecholamine hypersecretion or mass compression. Bladder paragangliomas are rare, especially those appearing after kidney transplantation. Here, we report a case of bladder paraganglioma developing after kidney transplantation. CASE SUMMARY: A 63-year-old woman received a kidney transplant 12 years ago and took oral immunosuppressants (cyclosporine, mizoribine, and methylprednisolone) for regular post-transplant treatment. The patient felt no discomfort and she came to the hospital for a routine checkup. A mass located in the bladder was incidentally discovered by computed tomography, and she underwent surgical treatment. A 2 cm × 2 cm invasive mass was found in the trigone of the bladder and the mass was removed. The diagnosis of paraganglioma was confirmed by morphology and immunophenotyping. The patient had a good prognosis and is still alive. CONCLUSION: Paraganglioma can grow in the bladder, which might cause no clinical symptoms. The diagnosis mainly depends on morphology and immunophenotyping. Surgical resection is an important treatment option for such patients.

Weighted Generalized Cross-Validation-Based Regularization for Broad Learning System.

The broad learning system (BLS) is an emerging flat network, which has demonstrated its outstanding performance in classification and regression problems. The regularization plays an important role in the performance of the BLS. In real applications, since the BLS network is usually expanded dynamically, a predetermined regularization parameter may reduce the performance of the network. Using a fixed regularization in some cases, the classification accuracy of the BLS decreases dramatically when we expand the network. To alleviate this problem, we propose a method that automatically finds appropriate regularization parameters for different datasets, which is based on the weighted generalized cross-validation (WGCV). The experimental results indicate that the WGCV method improves the performance of the BLS, and alleviates the accuracy decrease of the incremental learning algorithm.

Frequency Principle in Broad Learning System.

Deep neural networks have achieved breakthrough improvement in various application fields. Nevertheless, they usually suffer from a time-consuming training process because of the complicated structures of neural networks with a huge number of parameters. As an alternative, a fast and efficient discriminative broad learning system (BLS) is proposed, which takes the advantages of flat structure and incremental learning. The BLS has achieved outstanding performance in classification and regression problems. However, the previous studies ignored the reason why the BLS can generalize well. In this article, we focus on the interpretation from the viewpoint of the frequency domain. We discover the existence of the frequency principle in BLS, i.e., the BLS preferentially captures low-frequency components quickly and then fits the high frequencies during the incremental process of adding feature nodes and enhancement nodes. The frequency principle may be of great inspiration for expanding the application of BLS.

Current indications for endoscopic submucosal dissection of early gastric cancer.

The development of endoscopic treatment technology has further promoted the minimally invasive treatment of early gastric cancer (EGC). Endoscopic treatment has achieved better therapeutic effects in terms of safety and prognosis and is the preferred treatment method for patients who meet the indications for endoscopic treatment. However, the consequent problem is that some patients receiving endoscopic treatment may undergo non-curative resection, and the principle of follow-up management for non-curative resection patients deserves further attention. In addition, there are still debates on how to improve the accuracy of clinical staging, select a reasonable treatment method for patients who meet the expanded indications for endoscopic treatment, manage patients with positive endoscopic surgical margins, conduct research on function-preserving surgery, and manage the treatment of EGC under the current situation in China. Consequently, we aim to review current indications for endoscopic submucosal dissection of EGC in order to better inform treatment options.

The Correlation Between Mutant p53 Protein Expression and Cell Atypia in Early Differentiated Gastric Adenocarcinoma.

OBJECTIVE: This study aims to explore the correlation between the expression of mutant p53 protein and cellular atypia in early differentiated gastric adenocarcinoma (DGA). METHODS: A total of 107 cases of early DGA samples resected by endoscopic submucosal dissection (ESD) were collected from the Pathology Department of Beijing Friendship Hospital from January 2018 to December 2019. The EnVision two-step immunohistochemical method was used to detect the expression of mutant p53 protein in these cancer tissues, and the correlation with cell atypia was analyzed. RESULTS: In early DGA tissues, the expression rate of mutant p53 protein was significantly higher than in normal gastric mucosa (P < 0.01). However, the expression of mutant p53 protein was not correlated to age or gender (P > 0.05) but to the location of the tumor, depth of invasion, and degree of differentiation (P < 0.01). The expression of mutant p53 protein was closely correlated to cell atypia. Furthermore, this was weakly positive in low-grade atypical adenocarcinoma but strongly positive or negative in high-grade atypical adenocarcinoma, and there was a significant difference between these two (P < 0.01). CONCLUSION: Mutant p53 protein is highly expressed in early DGA, which can be used as an auxiliary index for the diagnosis of early gastric cancer. The different expression patterns of mutant p53 protein in high-grade and low-grade atypical gastric cancers suggest that these may have different genetic changes.

Basis Function Matrix-Based Flexible Coefficient Autoregressive Models: A Framework for Time Series and Nonlinear System Modeling.

We propose, in this paper, a framework for time series and nonlinear system modeling, called the basis function matrix-based flexible coefficient autoregressive (BFM-FCAR) model. It has very flexible nonlinear structure. We show that many famous nonlinear time series models can be derived under this framework by choosing the proper basis function matrices. Some probabilistic properties (the conditions of geometrical ergodicity) of the BFM-FCAR model are investigated. Taking advantage of the model structure, we present an efficient parameter estimation algorithm for the proposed framework by using the variable projection method. Finally, we show how new models are generated from the proposed framework.

Insights Into Algorithms for Separable Nonlinear Least Squares Problems.

Separable nonlinear least squares (SNLLS) problems have attracted interest in a wide range of research fields such as machine learning, computer vision, and signal processing. During the past few decades, several algorithms, including the joint optimization algorithm, alternated least squares (ALS) algorithm, embedded point iterations (EPI) algorithm, and variable projection (VP) algorithms, have been employed for solving SNLLS problems in the literature. The VP approach has been proven to be quite valuable for SNLLS problems and the EPI method has been successful in solving many computer vision tasks. However, no clear explanations about the intrinsic relationships of these algorithms have been provided in the literature. In this paper, we give some insights into these algorithms for SNLLS problems. We derive the relationships among different forms of the VP algorithms, EPI algorithm and ALS algorithm. In addition, the convergence and robustness of some algorithms are investigated. Moreover, the analysis of the VP algorithm generates a negative answer to Kaufman's conjecture. Numerical experiments on the image restoration task, fitting the time series data using the radial basis function network based autoregressive (RBF-AR) model, and bundle adjustment are given to compare the performance of different algorithms.

Recursive Variable Projection Algorithm for a Class of Separable Nonlinear Models.

In this article, we study the recursive algorithms for a class of separable nonlinear models (SNLMs) in which the parameters can be partitioned into a linear part and a nonlinear part. Such models are very common in machine learning, system identification, and signal processing. Utilizing the special structure of the SNLMs, we propose a recursive variable projection (RVP) algorithm, in which at each recursion, the linear parameters of the model are eliminated, and the nonlinear parameters are updated by the recursive Levenberg-Marquart algorithm. Then, based on the updated nonlinear parameters, the linear parameters are updated by the recursive least-squares algorithm. According to a convergence analysis of the RVP algorithm, the parameter estimation error is mean-square bounded. Numerical examples confirm the satisfactory performance of the proposed algorithm.

Malignant tumors associated with Peutz-Jeghers syndrome: Five cases from a single surgical unit.

BACKGROUND: Peutz-Jeghers syndrome (PJS) is an autosomal dominant inherited disease easily causing secondary malignant changes without effective treatments. AIM: To assess the clinical characteristics, diagnosis, and treatment of malignant changes secondary to PJS. METHODS: The clinical data of five patients with malignant changes secondary to PJS diagnosed and treated at Beijing Friendship Hospital from June 2014 to January 2017 were retrospectively analyzed; the follow-up ended in May 2018. RESULTS: There were three male and two female patients with an average age of 43.6 years. Intestinal obstruction, intussusception, and abdominal pain were the first symptoms. Computed tomography and gastrointestinal imaging combined with endoscopy helped evaluate the depth of tumor infiltration and determine the need for radical resection. Three patients underwent surgery. Postoperative pathology confirmed adenocarcinoma, genetic test indicated STK11 mutation, and the patients received chemotherapy, including one who succumbed to tumor progression 6 months post-surgery. Other two patients underwent endoscopic resection, and postoperative pathology confirmed high grade intraepithelial neoplasia. The surviving patients had no recurrence by May 2018. CONCLUSION: Endoscopy combined with computed tomography and gastrointestinal imaging is of great significance in the diagnosis and treatment of PJS, and pathological examination and gene detection are the gold standards for detecting malignant changes secondary to PJS. Some malignant polyps can be removed under endoscopy, and surgery is feasible when malignant polyps cannot be removed under an endoscope. For patients unable to achieve R0 resection, clinical symptoms should be relieved, and postoperative adjuvant chemotherapy could improve long-term prognosis. Meanwhile, close and regular surveillance should be conducted to prevent severe complications.

Term Selection for a Class of Separable Nonlinear Models.

In this paper, we consider the term selection problem for a class of separable nonlinear models. The strategy is a two-step process in which the nonlinear parameters of the model are first optimized by a variable projection method, and then the least absolute shrinkage and selection operator are adopted to obtain a sparse solution by picking out the critical terms automatically. This process may be repeated several times. The proposed algorithm is tested on parameter estimation problems for an exponential model and a neural network-based model. The numerical results show that the proposed algorithm can pick out the appropriate terms from the overparameterized model and the obtained parsimonious model performs better than other methods.

Modified Gram-Schmidt Method-Based Variable Projection Algorithm for Separable Nonlinear Models.

Separable nonlinear models are very common in various research fields, such as machine learning and system identification. The variable projection (VP) approach is efficient for the optimization of such models. In this paper, we study various VP algorithms based on different matrix decompositions. Compared with the previous method, we use the analytical expression of the Jacobian matrix instead of finite differences. This improves the efficiency of the VP algorithms. In particular, based on the modified Gram-Schmidt (MGS) method, a more robust implementation of the VP algorithm is introduced for separable nonlinear least-squares problems. In numerical experiments, we compare the performance of five different implementations of the VP algorithm. Numerical results show the efficiency and robustness of the proposed MGS method-based VP algorithm.

Spontaneous cerebrospinal fluid rhinorrhea: A case report and analysis.

INTRODUCTION: Spontaneous cerebrospinal fluid leakage is usually caused by developmental abnormalities and is rare, accounting for approximately 5% of the cases of cerebrospinal fluid (CSF) leakage. To the best of our knowledge, clival dysplasia-caused CSF rhinorrhea has never been reported in the neurosurgical field. CONCLUSION: Spontaneous cerebrospinal fluid rhinorrhea is often treated by surgery, and a transsphenoidal approach repair is the main surgical method used, offering the advantages of less trauma, fewer complications, rapid postoperative recovery, and low recurrence rate.

Efficacy and safety of using premedication with simethicone/Pronase during upper gastrointestinal endoscopy examination with sedation: A single center, prospective, single blinded, randomized controlled trial.

BACKGROUND AND AIM: To investigate the efficacy and safety of premedication with simethicone/Pronase during esophagogastroduodenoscopy (EGD) with sedation. METHODS: Six hundred and ten patients were randomly allocated to two groups based on type of premedication given. Premedication used in the control group was 10 mL lidocaine hydrochloride mucilage (LHM, N = 314) and premedication used in the intervention group was 80 mL simethicone/Pronase solution plus 10 mL lidocaine hydrochloride mucilage (SP/LHM, N = 296). EGD was done under sedation. Visibility scores, number of mucosal areas that needed cleansing, water consumption for cleansing, time taken for examination, diminutive lesions, pathological diagnosis, patients' gag reflex and oxygenation (pulse oximetry) were recorded. RESULTS: SP/LHM has significantly lower total visibility score than LHM (7.978 ± 1.526 vs 6.348 ± 1.097, P < 0.01). During the procedure, number of intragastric areas that needed cleansing and amount of water consumed were significantly less in the SP/LHM than in the LHM group (P < 0.01). In SP/LHM (P = 0.01), endoscopy procedure duration was significantly longer. Although there was no significant difference in rate of detection of diminutive lesions between LHM and SP/LHM, the endoscopist carried out more biopsies in SP/LHM. This led to a higher rate of diagnosis of atrophic gastritis (P = 0.014) and intestinal metaplasia (P = 0.024). There was no significant difference in gag reflex (P = 0.604) and oxygenation during the endoscopy procedure for either group of patients. CONCLUSION: Routine use of premedication with simethicone/Pronase should be recommended during EGD with sedation.

Evaluation and identification of factors related to KRAS and BRAF gene mutations in colorectal cancer: A meta-analysis.

OBJECTIVE: The aim of this meta-analysis is to evaluate the distribution pattern of KRAS and BRAF mutations in colorectal cancer. MATERIALS AND METHODS: The database was searched without language restrictions. Meta-analyses were conducted using the STATA software. We calculated the odds ratio (OR) and its 95% confidence interval (95% CI) to estimate the distribution of and correlation between KRAS and BRAF mutations, CpG island methylator phenotype (CIMP), and microsatellite instability (MSI) in left- and right-sided colorectal cancer. RESULTS: The studies were divided into five groups: (1) distribution of KRAS/BRAF mutations in distal and proximal colorectal cancer, the summary OR value was 1.24 versus 4.03, (2) distribution of KRAS/BRAF mutations in CIMP-low/Neg and CIMP-high (CIMP-H) tumors, the summary OR value was 0.77 versus 10.49, (3) distribution of KRAS/BRAF mutations in MSI-low (MSI-L)/microsatellite stable (MSS) and MSI-high (MSI-H) tumors, the summary OR value was 0.51 versus 9.60, (4) proportion of CIMP-H/MSI-H tumors among distal and proximal colorectal tumors, the summary OR value was 3.66 versus 6.54, and (5) proportion of CIMP-H tumors among MSI-L/MSS and MSI-H tumors, the summary OR value was 5.87. CONCLUSION: The meta-analysis reveals that KRAS has a slightly higher mutation rate in MSI-L/MSS tumors. Moreover, BRAF mutations have higher detection rates in right-sided colorectal cancer, which suggests that BRAF mutations are likely in CIMP-H tumors. Therefore, based on these findings, the molecular diagnostic tests to be conducted in colorectal cancer patients can be determined according to the location/clinical features of the tumor.