Sterigmatocystin induces autophagic and apoptotic cell death of liver cancer cells via downregulation of XIAP.

XIAP, or the X-linked Inhibitor of Apoptosis Protein, is the most extensively studied member within the IAP gene family. It possesses the capability to impede apoptosis through direct inhibition of caspase activity. Various kinds of cancers overexpress XIAP to enable cancer cells to avoid apoptosis. Consequently, the inhibition of XIAP holds significant clinical implications for the development of anti-tumor medications and the treatment of cancer. In this study, sterigmatocystin, a natural compound obtained from the genus asperigillus, was demonstrated to be able to induce apoptotic and autophagic cell death in liver cancer cells. Mechanistically, sterigmatocystin induces apoptosis by downregulation of XIAP expression. Additionally, sterigmatocystin treatment induces cell cycle arrest, blocks cell proliferation, and slows down colony formation in liver cancer cells. Importantly, sterigmatocystin exhibits a remarkable therapeutic effect in a nude mice model. Our findings revealed a novel mechanism through which sterigmatocystin induces apoptotic and autophagic cell death of liver cancer cells by suppressing XIAP expression, this offers a promising therapeutic approach for treating liver cancer patients.

RDFC-GAN: RGB-Depth Fusion CycleGAN for Indoor Depth Completion.

Raw depth images captured in indoor scenarios frequently exhibit extensive missing values due to the inherent limitations of the sensors and environments. For example, transparent materials frequently elude detection by depth sensors; surfaces may introduce measurement inaccuracies due to their polished textures, extended distances, and oblique incidence angles from the sensor. The presence of incomplete depth maps imposes significant challenges for subsequent vision applications, prompting the development of numerous depth completion techniques to mitigate this problem. Numerous methods excel at reconstructing dense depth maps from sparse samples, but they often falter when faced with extensive contiguous regions of missing depth values, a prevalent and critical challenge in indoor environments. To overcome these challenges, we design a novel two-branch end-to-end fusion network named RDFC-GAN, which takes a pair of RGB and incomplete depth images as input to predict a dense and completed depth map. The first branch employs an encoder-decoder structure, by adhering to the Manhattan world assumption and utilizing normal maps from RGB-D information as guidance, to regress the local dense depth values from the raw depth map. The other branch applies an RGB-depth fusion CycleGAN, adept at translating RGB imagery into detailed, textured depth maps while ensuring high fidelity through cycle consistency. We fuse the two branches via adaptive fusion modules named W-AdaIN and train the model with the help of pseudo depth maps. Comprehensive evaluations on NYU-Depth V2 and SUN RGB-D datasets show that our method significantly enhances depth completion performance particularly in realistic indoor settings.

Distributional generative adversarial imitation learning with reproducing kernel generalization.

Generative adversarial imitation learning (GAIL) regards imitation learning (IL) as a distribution matching problem between the state-action distributions of the expert policy and the learned policy. In this paper, we focus on the generalization and computational properties of policy classes. We prove that the generalization can be guaranteed in GAIL when the class of policies is well controlled. With the capability of policy generalization, we introduce distributional reinforcement learning (RL) into GAIL and propose the greedy distributional soft gradient (GDSG) algorithm to solve GAIL. The main advantages of GDSG can be summarized as: (1) Q-value overestimation, a crucial factor leading to the instability of GAIL with off-policy training, can be alleviated by distributional RL. (2) By considering the maximum entropy objective, the policy can be improved in terms of performance and sample efficiency through sufficient exploration. Moreover, GDSG attains a sublinear convergence rate to a stationary solution. Comprehensive experimental verification in MuJoCo environments shows that GDSG can mimic expert demonstrations better than previous GAIL variants.

CATRO: Channel Pruning via Class-Aware Trace Ratio Optimization.

Deep convolutional neural networks are shown to be overkill with high parametric and computational redundancy in many application scenarios, and an increasing number of works have explored model pruning to obtain lightweight and efficient networks. However, most existing pruning approaches are driven by empirical heuristics and rarely consider the joint impact of channels, leading to unguaranteed and suboptimal performance. In this article, we propose a novel channel pruning method via class-aware trace ratio optimization (CATRO) to reduce the computational burden and accelerate the model inference. Utilizing class information from a few samples, CATRO measures the joint impact of multiple channels by feature space discriminations and consolidates the layerwise impact of preserved channels. By formulating channel pruning as a submodular set function maximization problem, CATRO solves it efficiently via a two-stage greedy iterative optimization procedure. More importantly, we present theoretical justifications on convergence of CATRO and performance of pruned networks. Experimental results demonstrate that CATRO achieves higher accuracy with similar computation cost or lower computation cost with similar accuracy than other state-of-the-art channel pruning algorithms. In addition, because of its class-aware property, CATRO is suitable to prune efficient networks adaptively for various classification subtasks, enhancing handy deployment and usage of deep networks in real-world applications.

DAIS: Automatic Channel Pruning via Differentiable Annealing Indicator Search.

The convolutional neural network (CNN) has achieved great success in fulfilling computer vision tasks despite large computation overhead against efficient deployment. Channel pruning is usually applied to reduce the model redundancy while preserving the network structure, such that the pruned network can be easily deployed in practice. However, existing channel pruning methods require hand-crafted rules, which can result in a degraded model performance with respect to the tremendous potential pruning space given large neural networks. In this article, we introduce differentiable annealing indicator search (DAIS) that leverages the strength of neural architecture search in the channel pruning and automatically searches for the effective pruned model with given constraints on computation overhead. Specifically, DAIS relaxes the binarized channel indicators to be continuous and then jointly learns both indicators and model parameters via bi-level optimization. To bridge the non-negligible discrepancy between the continuous model and the target binarized model, DAIS proposes an annealing-based procedure to steer the indicator convergence toward binarized states. Moreover, DAIS designs various regularizations based on a priori structural knowledge to control the pruning sparsity and to improve model performance. Experimental results show that DAIS outperforms state-of-the-art pruning methods on CIFAR-10, CIFAR-100, and ImageNet.

Mesenchymal stem/stromal cells in the pathogenesis and regenerative therapy of inflammatory bowel diseases.

Inflammatory bowel diseases (IBDs) represent a group of chronic inflammatory disorders of the gastrointestinal (GI) tract including ulcerative colitis (UC), Crohn's disease (CD), and unclassified IBDs. The pathogenesis of IBDs is related to genetic susceptibility, environmental factors, and dysbiosis that can lead to the dysfunction of immune responses and dysregulated homeostasis of local mucosal tissues characterized by severe inflammatory responses and tissue damage in GI tract. To date, extensive studies have indicated that IBDs cannot be completely cured and easy to relapse, thus prompting researchers to find novel and more effective therapeutics for this disease. Due to their potent multipotent differentiation and immunomodulatory capabilities, mesenchymal stem/stromal cells (MSCs) not only play an important role in regulating immune and tissue homeostasis but also display potent therapeutic effects on various inflammatory diseases, including IBDs, in both preclinical and clinical studies. In this review, we present a comprehensive overview on the pathological mechanisms, the currently available therapeutics, particularly, the potential application of MSCs-based regenerative therapy for IBDs.

Gingiva-Derived Mesenchymal Stem Cells Attenuate Imiquimod- (IMQ-) Induced Murine Psoriasis-Like Skin Inflammation.

Human gingiva-derived mesenchymal stem cells (GMSCs) are isolated from the gingival propria with promising regenerative, immunomodulatory, and anti-inflammatory properties. Recently, several studies, including ours, have found that GMSCs have the therapeutic potentials of nerve regeneration and skin disorders in various types such as the cell itself, cell-free conditioned medium, or extracellular vesicles (EVs). However, the mechanobiological behavior of GMSCs is closely related to the culture conditions. Therefore, the purpose of this study was to evaluate the function of human GMSCs on imiquimod- (IMQ-) induced murine psoriasis-like skin inflammation in two-dimensional (2D) and three-dimensional (3D) culture conditions. Here, we isolated and characterized GMSCs in 2D and 3D culture conditions and found that GMSCs in 2D and 3D infusion can significantly ameliorate the IMQ-induced murine psoriasis-like skin inflammation, reduce the levels of Th1- and Th17-related cytokines IFN-γ, TNF-α, IL-6, IL-17A, IL-17F, IL-21, and IL-22, and upregulate the percentage of spleen CD25+CD3+ T cells while downregulate the percentage of spleen IL-17+CD3+ T cells. In summary, our novel findings reveal that GMSCs in 2D and 3D infusion may possess therapeutic effects in the treatment of psoriasis.

Robust Unsupervised Video Anomaly Detection by Multipath Frame Prediction.

Video anomaly detection is commonly used in many applications, such as security surveillance, and is very challenging. A majority of recent video anomaly detection approaches utilize deep reconstruction models, but their performance is often suboptimal because of insufficient reconstruction error differences between normal and abnormal video frames in practice. Meanwhile, frame prediction-based anomaly detection methods have shown promising performance. In this article, we propose a novel and robust unsupervised video anomaly detection method by frame prediction with a proper design which is more in line with the characteristics of surveillance videos. The proposed method is equipped with a multipath ConvGRU-based frame prediction network that can better handle semantically informative objects and areas of different scales and capture spatial-temporal dependencies in normal videos. A noise tolerance loss is introduced during training to mitigate the interference caused by background noise. Extensive experiments have been conducted on the CUHK Avenue, ShanghaiTech Campus, and UCSD Pedestrian datasets, and the results show that our proposed method outperforms existing state-of-the-art approaches. Remarkably, our proposed method obtains the frame-level AUROC score of 88.3% on the CUHK Avenue dataset.

Lactobacillus paracasei R3 protects against dextran sulfate sodium (DSS)-induced colitis in mice via regulating Th17/Treg cell balance.

Inflammatory bowel diseases (IBD), mainly comprising ulcerative colitis (UC) and Crohn's Disease, are most often a polygenic disorder with contributions from the intestinal microbiome, defects in barrier function, and dysregulated host responses to microbial stimulation. Strategies that target the microbiota have emerged as potential therapies and, of these, probiotics have gained the greatest attention. Herein, we isolated a strain of Lactobacillus paracasei R3 (L.p R3) with strong biofilm formation ability from infant feces. Interestingly, we also found L.p R3 strain can ameliorate the general symptoms of murine colitis, alleviate inflammatory cell infiltration and inhibit Th17 while promote Treg function in murine dextran sulfate sodium (DSS)-induced colitis. Overall, this study suggested that L.p R3 strain significantly improves the symptoms and the pathological damage of mice with colitis and influences the immune function by regulating Th17/Treg cell balance in DSS-induced colitis in mice.

Mesenchymal Stem Cells and Tuberculosis: Clinical Challenges and Opportunities.

Tuberculosis (TB) is one of the communicable diseases caused by Mycobacterium tuberculosis (Mtb) infection, affecting nearly one-third of the world's population. However, because the pathogenesis of TB is still not fully understood and the development of anti-TB drug is slow, TB remains a global public health problem. In recent years, with the gradual discovery and confirmation of the immunomodulatory properties of mesenchymal stem cells (MSCs), more and more studies, including our team's research, have shown that MSCs seem to be closely related to the growth status of Mtb and the occurrence and development of TB, which is expected to bring new hope for the clinical treatment of TB. This article reviews the relationship between MSCs and the occurrence and development of TB and the potential application of MSCs in the treatment of TB.

Beta-1 syntrophin (SNTB1) regulates colorectal cancer progression and stemness via regulation of the Wnt/ß-catenin signaling pathway.

BACKGROUND: Beta-1 syntrophin (SNTB1) is an intracellular scaffold protein that provides a platform for the formation of signal transduction complexes, thereby modulating and coordinating various intracellular signaling events and crucial cellular processes. However, the physiological role of SNTB1 is poorly understood. This study aims to explore the role of SNTB1 in colorectal cancer (CRC) tumorigenesis and progression, with particular focus on SNTB1's expression pattern, clinical relevance, and possible molecular mechanism in CRC development. METHODS: SNTB1 expression was analyzed in both clinical tissues and The Cancer Genome Atlas (TCGA) database. Real-time polymerase chain reaction (PCR), Western blot, and immunohistochemical assays were used to detect the relative mRNA and protein levels of SNTB1. Statistical analysis was performed to examine the correlation between SNTB1 expression and the clinicopathological characteristics of patients with CRC. Bioinformatics gene set enrichment analysis (GSEA), Western blot, luciferase assay, and agonist recovery assays were conducted to evaluate the relevance of SNTB1 and the ß-catenin signaling pathway in CRC. A flow cytometry-based Hoechst 33342 efflux assay was applied to assess the proportion of the side population (SP) within total CRC cells. RESULTS: Elevated levels of SNTB1 were identified in CRC tissues and cell lines. The elevation of SNTB1 was positively correlated with the degree of malignancy and poor prognosis in CRC. We further revealed that, by modulating the ß-catenin signaling pathway, silencing SNTB1 expression suppressed tumor growth and cancer stemness in vitro, as well as tumorigenesis in vivo. CONCLUSIONS: These findings suggest that SNTB1 plays a crucial role in colorectal tumorigenesis and progression by modulating ß-catenin signaling and the stemness maintenance of cancer cells.

Ectodysplasin A receptor (EDAR) promotes colorectal cancer cell proliferation via regulation of the Wnt/ß-catenin signaling pathway.

Colorectal cancer is the second leading cause of cancer mortality worldwide with poor prognosis and high recurrence. Aberrant Wnt/ß-catenin signaling promotes oncogenesis by transcriptional activation of c-Myc and its downstream signals. EDAR is characterized as an important effector of canonical Wnt signaling in developing skin appendages, but the interplay between EDAR and Wnt signaling in tumorigenesis and progression remains to be elucidated. In this study, we revealed that EDAR expression is prevalently elevated in colorectal cancer tissues compared with normal tissues. Further analysis suggests there is a strict correlation between EDAR expression and colorectal cancer progression. EDAR silencing by shRNA in colorectal cancer cells showed proliferative suppression via retarding cell cycle at G1 phase. Xenograft mice transplanted with shEDAR-transduced tumor cells significantly alleviated tumor burden in comparison with control mice. Furthermore, downregulation of EDAR was accompanied by reduction of ß-catenin, c-Myc and other G1 cell cycle regulators, while ß-catenin agonist restored the expression of these proteins and overrode the proliferative block induced by EDAR knockdown. These findings indicate that EDAR functions as a component of Wnt/ß-catenin signaling pathway, and is a potential modulator in colorectal carcinogenesis.

Benchmarking deep learning models on large healthcare datasets.

Deep learning models (aka Deep Neural Networks) have revolutionized many fields including computer vision, natural language processing, speech recognition, and is being increasingly used in clinical healthcare applications. However, few works exist which have benchmarked the performance of the deep learning models with respect to the state-of-the-art machine learning models and prognostic scoring systems on publicly available healthcare datasets. In this paper, we present the benchmarking results for several clinical prediction tasks such as mortality prediction, length of stay prediction, and ICD-9 code group prediction using Deep Learning models, ensemble of machine learning models (Super Learner algorithm), SAPS II and SOFA scores. We used the Medical Information Mart for Intensive Care III (MIMIC-III) (v1.4) publicly available dataset, which includes all patients admitted to an ICU at the Beth Israel Deaconess Medical Center from 2001 to 2012, for the benchmarking tasks. Our results show that deep learning models consistently outperform all the other approaches especially when the 'raw' clinical time series data is used as input features to the models.

Recurrent Neural Networks for Multivariate Time Series with Missing Values.

Multivariate time series data in practical applications, such as health care, geoscience, and biology, are characterized by a variety of missing values. In time series prediction and other related tasks, it has been noted that missing values and their missing patterns are often correlated with the target labels, a.k.a., informative missingness. There is very limited work on exploiting the missing patterns for effective imputation and improving prediction performance. In this paper, we develop novel deep learning models, namely GRU-D, as one of the early attempts. GRU-D is based on Gated Recurrent Unit (GRU), a state-of-the-art recurrent neural network. It takes two representations of missing patterns, i.e., masking and time interval, and effectively incorporates them into a deep model architecture so that it not only captures the long-term temporal dependencies in time series, but also utilizes the missing patterns to achieve better prediction results. Experiments of time series classification tasks on real-world clinical datasets (MIMIC-III, PhysioNet) and synthetic datasets demonstrate that our models achieve state-of-the-art performance and provide useful insights for better understanding and utilization of missing values in time series analysis.

Utilizing Machine Learning and Automated Performance Metrics to Evaluate Robot-Assisted Radical Prostatectomy Performance and Predict Outcomes.

PURPOSE: Surgical performance is critical for clinical outcomes. We present a novel machine learning (ML) method of processing automated performance metrics (APMs) to evaluate surgical performance and predict clinical outcomes after robot-assisted radical prostatectomy (RARP). MATERIALS AND METHODS: We trained three ML algorithms utilizing APMs directly from robot system data (training material) and hospital length of stay (LOS; training label) (≤2 days and >2 days) from 78 RARP cases, and selected the algorithm with the best performance. The selected algorithm categorized the cases as "Predicted as expected LOS (pExp-LOS)" and "Predicted as extended LOS (pExt-LOS)." We compared postoperative outcomes of the two groups (Kruskal-Wallis/Fisher's exact tests). The algorithm then predicted individual clinical outcomes, which we compared with actual outcomes (Spearman's correlation/Fisher's exact tests). Finally, we identified five most relevant APMs adopted by the algorithm during predicting. RESULTS: The "Random Forest-50" (RF-50) algorithm had the best performance, reaching 87.2% accuracy in predicting LOS (73 cases as "pExp-LOS" and 5 cases as "pExt-LOS"). The "pExp-LOS" cases outperformed the "pExt-LOS" cases in surgery time (3.7 hours vs 4.6 hours, p = 0.007), LOS (2 days vs 4 days, p = 0.02), and Foley duration (9 days vs 14 days, p = 0.02). Patient outcomes predicted by the algorithm had significant association with the "ground truth" in surgery time (p < 0.001, r = 0.73), LOS (p = 0.05, r = 0.52), and Foley duration (p < 0.001, r = 0.45). The five most relevant APMs, adopted by the RF-50 algorithm in predicting, were largely related to camera manipulation. CONCLUSION: To our knowledge, ours is the first study to show that APMs and ML algorithms may help assess surgical RARP performance and predict clinical outcomes. With further accrual of clinical data (oncologic and functional data), this process will become increasingly relevant and valuable in surgical assessment and training.

Deep Learning Solutions for Classifying Patients on Opioid Use.

Opioid analgesics, as commonly prescribed medications used for relieving pain in patients, are especially prevalent in US these years. However, an increasing amount of opioid misuse and abuse have caused lots of consequences. Researchers and clinicians have attempted to discover the factors leading to opioid long-term use, dependence, and abuse, but only limited incidents are understood from previous works. Motivated by recent successes of deep learning and the abundant amount of electronic health records, we apply state-of-the-art deep and recurrent neural network models on a dataset of more than one hundred thousand opioid users. Our models are shown to achieve robust and superior results on classifying opioid users, and are able to extract key factors for different opioid user groups. This work is also a good demonstration on adopting novel deep learning methods for real-world health care problems.

Interpretable Deep Models for ICU Outcome Prediction.

Exponential surge in health care data, such as longitudinal data from electronic health records (EHR), sensor data from intensive care unit (ICU), etc., is providing new opportunities to discover meaningful data-driven characteristics and patterns ofdiseases. Recently, deep learning models have been employedfor many computational phenotyping and healthcare prediction tasks to achieve state-of-the-art performance. However, deep models lack interpretability which is crucial for wide adoption in medical research and clinical decision-making. In this paper, we introduce a simple yet powerful knowledge-distillation approach called interpretable mimic learning, which uses gradient boosting trees to learn interpretable models and at the same time achieves strong prediction performance as deep learning models. Experiment results on Pediatric ICU dataset for acute lung injury (ALI) show that our proposed method not only outperforms state-of-the-art approaches for morality and ventilator free days prediction tasks but can also provide interpretable models to clinicians.

Causal Phenotype Discovery via Deep Networks.

The rapid growth of digital health databases has attracted many researchers interested in using modern computational methods to discover and model patterns of health and illness in a research program known as computational phenotyping. Much of the work in this area has focused on traditional statistical learning paradigms, such as classification, prediction, clustering, pattern mining. In this paper, we propose a related but different paradigm called causal phenotype discovery, which aims to discover latent representations of illness that are causally predictive. We illustrate this idea with a two-stage framework that combines the latent representation learning power of deep neural networks with state-of-the-art tools from causal inference. We apply this framework to two large ICU time series data sets and show that it can learn features that are predictively useful, that capture complex physiologic patterns associated with critical illnesses, and that are potentially more clinically meaningful than manually designed features.