Self-Interested Coalitional Crowdsensing for Multi-Agent Interactive Environment Monitoring.

As a promising paradigm, mobile crowdsensing (MCS) takes advantage of sensing abilities and cooperates with multi-agent reinforcement learning technologies to provide services for users in large sensing areas, such as smart transportation, environment monitoring, etc. In most cases, strategy training for multi-agent reinforcement learning requires substantial interaction with the sensing environment, which results in unaffordable costs. Thus, environment reconstruction via extraction of the causal effect model from past data is an effective way to smoothly accomplish environment monitoring. However, the sensing environment is often so complex that the observable and unobservable data collected are sparse and heterogeneous, affecting the accuracy of the reconstruction. In this paper, we focus on developing a robust multi-agent environment monitoring framework, called self-interested coalitional crowdsensing for multi-agent interactive environment monitoring (SCC-MIE), including environment reconstruction and worker selection. In SCC-MIE, we start from a multi-agent generative adversarial imitation learning framework to introduce a new self-interested coalitional learning strategy, which forges cooperation between a reconstructor and a discriminator to learn the sensing environment together with the hidden confounder while providing interpretability on the results of environment monitoring. Based on this, we utilize the secretary problem to select suitable workers to collect data for accurate environment monitoring in a real-time manner. It is shown that SCC-MIE realizes a significant performance improvement in environment monitoring compared to the existing models.

Targeting DDX11 promotes PARP inhibitor sensitivity in hepatocellular carcinoma by attenuating BRCA2-RAD51 mediated homologous recombination.

Homologous recombination (HR) is a major DNA double-strand break (DSB) repair pathway of clinical interest because of treatment with poly(ADP-ribose) polymerase inhibitors (PARPi). Cooperation between RAD51 and BRCA2 is pivotal for DNA DSB repair, and its dysfunction induces HR deficiency and sensitizes cancer cells to PARPi. The depletion of the DEAD-box protein DDX11 was found to suppress HR in hepatocellular carcinoma (HCC) cells. The HR ability of HCC cells is not always dependent on the DDX11 level because of natural DDX11 mutations. In Huh7 cells, natural DDX11 mutations were detected, increasing the susceptibility of Huh7 cells to olaparib in vitro and in vivo. The HR deficiency of Huh7 cells was restored when CRISPR/Cas9-mediated knock-in genomic editing was used to revert the DDX11 Q238H mutation to wild type. The DDX11 Q238H mutation impeded the phosphorylation of DDX11 by ATM at serine 237, preventing the recruitment of RAD51 to damaged DNA sites by disrupting the interaction between RAD51 and BRCA2. Clinically, a high level of DDX11 correlated with advanced clinical characteristics and a poor prognosis and served as an independent risk factor for overall and disease-free survival in patients with HCC. We propose that HCC with a high level of wild-type DDX11 tends to be more resistant to PARPi because of enhanced recombination repair, and the key mutation of DDX11 (Q238H) is potentially exploitable.

Unveiling immune-metabolic interaction in Atherosclerosis via Comprehensive Landscape of Hub Genes and Immune Microenvironment.

OBJECTIVE: Atherosclerosis (AS) as a major cause of cardiovascular diseases, is considered a chronic inflammatory disease and accelerates by inflammation, lipid metabolism disorder and other mechanisms. AS pathogenesis and its relationship with immune regulation and metabolic interactions is still not fully elucidated. The purpose of this study is to delve into the correlation between mitochondrial metabolism and immunity in AS, and identify potential therapeutic targets for clinical treatment. METHODS: Hub genes associated with mitochondrial metabolism and the pathogenesis of AS were identified by performing differentially expressed genes (DEGs) analysis and Weighted Gene Co-expression Network Analysis (WGCNA) based on two gene expression datasets (GSE100927 and GSE43292). And the biological processes and pathways of DEGs were determined through gene ontology (GO) and Gene Set Enrichment Analysis (GSEA) analysis. Then stepwise regression, random forest, and Lasso regression machine learning were used to evaluate the diagnostic value of hub genes. After that, the immune infiltration and single cell sequencing dataset GSE184073 were analyzed, and the immune cell composition in peripheral blood from AS patients using Mass Cytometry were detected to further consider the influence of immunoregulation. RESULTS: Ten hub genes associated with mitochondrial metabolism and AS pathogenesis were identified, including NDUFS4, AIFM3, IDUA, TNF, CHKA, SLC11A1, SLC35C1, SLC37A2, ARSB, SLC16A5. GO and GSEA analysis showed their correlation with immunity and inflammation. Lasso regression revealed that TNF and ARSB had relatively good diagnostic performance. Further exploration was conducted on the expression of these hub genes in the immune microenvironment and their correlation with different immune factors. Mass cytometry demonstrated the influence of the vascular immune microenvironment on the pathogenesis of AS. CONCLUSIONS: Our study provides a more comprehensive understanding of the complex relationships between immune and metabolic factors and their impact on the microenvironment of AS. The identification of hub genes in AS may provide new targets for therapeutic intervention.

Suberoylanilide hydroxamic acid (SAHA) attenuates memory impairment in the offspring of rats exposed to sevoflurane anesthesia.

BACKGROUND: SAHA was reported to enhance the expression of miR-129-5p, which was predicted to bind to 3' UTR of CASP-6, a gene playing crucial roles in the pathogenesis of memory impairment. Whether SAHA/miR-129-5p/CASP-6 is involved in the pathogenesis of prenatal exposure to sevoflurane remains to be explored. METHODS: Morris water maze test was performed to evaluate the functional parameters of learning and memory. Quantitative real-time qPCR was carried out to analyze the expression of miRNAs and CASP-6 mRNA under different conditions. RESULTS: Sevoflurane exposure of pregnant rats and SAHA treatment of the offspring had no effect on the blood gases, litter size, survival rate and weight. SAHA administration remarkably reversed the learning and memory impairment in prenatal rats caused by sevoflurane exposure. Mechanistically, the abnormal expression of miR-129-5p and CASP-6 in the offspring of pregnant rats exposed to sevoflurane was effectively restored by SAHA treatment. The luciferase activity of CASP-6 vector was effectively inhibited by miR-129-5p in primary neuron cells of rats. Moreover, the expression of CASP-6 mRNA and protein was significantly suppressed by miR-129-5p and SAHA treatment in a dose-dependent manner. CONCLUSION: Our work demonstrated that the administration of SAHA suppressed the expression of CASP-6 via modulating the expression of miR-129-5p, and SAHA may rescue the apoptosis of neurons caused by exposure to sevoflurane. The underlying mechanism might be the ability of SAHA to relieve learning and memory impairment in the offspring of the pregnant rats exposed to sevoflurane.

Social distance "nudge:" a context aware mHealth intervention in response to COVID pandemics.

The impact of the COVID pandemic to our society is unprecedented in our time. As coronavirus mutates, maintaining social distance remains an essential step in defending personal as well as public health. This study conceptualizes the social distance "nudge" and explores the efficacy of mHealth digital intervention, while developing and validating a choice architecture that aims to influence users' behavior in maintaining social distance for their own self-interest. End-user nudging experiments were conducted via a mobile phone app that was developed as a research artifact. The accuracy of social distance nudging was validated in both United States and Japan. Future work will consider behavioral studies to better understand the effectiveness of this digital nudging intervention.

Assessing the Impact of Imputation on the Interpretations of Prediction Models: A Case Study on Mortality Prediction for Patients with Acute Myocardial Infarction.

Acute myocardial infarction poses significant health risks and financial burden on healthcare and families. Prediction of mortality risk among AM! patients using rich electronic health record (EHR) data can potentially save lives and healthcare costs. Nevertheless, EHR-based prediction models usually use a missing data imputation method without considering its impact on the performance and interpretability of the model, hampering its real-world applicability in the healthcare setting. This study examines the impact of different methods for imputing missing values in EHR data on both the performance and the interpretations of predictive models. Our results showed that a small standard deviation in root mean squared error across different runs of an imputation method does not necessarily imply a small standard deviation in the prediction models' performance and interpretation. We also showed that the level of missingness and the imputation method used can have a significant impact on the interpretation of the models.

DeepConsensus: Consensus-based Interpretable Deep Neural Networks with Application to Mortality Prediction.

Deep neural networks have achieved remarkable success in various challenging tasks. However, the black-box nature of such networks is not acceptable to critical applications, such as healthcare. In particular, the existence of adversarial examples and their overgeneralization to irrelevant, out-of-distribution inputs with high confidence makes it difficult, if not impossible, to explain decisions by such networks. In this paper, we analyze the underlying mechanism of generalization of deep neural networks and propose an (n, k) consensus algorithm which is insensitive to adversarial examples and can reliably reject out-of-distribution samples. Furthermore, the consensus algorithm is able to improve classification accuracy by using multiple trained deep neural networks. To handle the complexity of deep neural networks, we cluster linear approximations of individual models and identify highly correlated clusters among different models to capture feature importance robustly, resulting in improved interpretability. Motivated by the importance of building accurate and interpretable prediction models for healthcare, our experimental results on an ICU dataset show the effectiveness of our algorithm in enhancing both the prediction accuracy and the interpretability of deep neural network models on one-year patient mortality prediction. In particular, while the proposed method maintains similar interpretability as conventional shallow models such as logistic regression, it improves the prediction accuracy significantly.

Explainable artificial intelligence models using real-world electronic health record data: a systematic scoping review.

OBJECTIVE: To conduct a systematic scoping review of explainable artificial intelligence (XAI) models that use real-world electronic health record data, categorize these techniques according to different biomedical applications, identify gaps of current studies, and suggest future research directions. MATERIALS AND METHODS: We searched MEDLINE, IEEE Xplore, and the Association for Computing Machinery (ACM) Digital Library to identify relevant papers published between January 1, 2009 and May 1, 2019. We summarized these studies based on the year of publication, prediction tasks, machine learning algorithm, dataset(s) used to build the models, the scope, category, and evaluation of the XAI methods. We further assessed the reproducibility of the studies in terms of the availability of data and code and discussed open issues and challenges. RESULTS: Forty-two articles were included in this review. We reported the research trend and most-studied diseases. We grouped XAI methods into 5 categories: knowledge distillation and rule extraction (N = 13), intrinsically interpretable models (N = 9), data dimensionality reduction (N = 8), attention mechanism (N = 7), and feature interaction and importance (N = 5). DISCUSSION: XAI evaluation is an open issue that requires a deeper focus in the case of medical applications. We also discuss the importance of reproducibility of research work in this field, as well as the challenges and opportunities of XAI from 2 medical professionals' point of view. CONCLUSION: Based on our review, we found that XAI evaluation in medicine has not been adequately and formally practiced. Reproducibility remains a critical concern. Ample opportunities exist to advance XAI research in medicine.

The Enigma of Centriole Loss in the 1182-4 Cell Line.

The Drosophilamelanogaster cell line 1182-4, which constitutively lacks centrioles, was established many years ago from haploid embryos laid by females homozygous for the maternal haploid (mh) mutation. This was the first clear example of animal cells regularly dividing in the absence of this organelle. However, the cause of the acentriolar nature of the 1182-4 cell line remained unclear and could not be clearly assigned to a particular genetic event. Here, we detail historically the longstanding mystery of the lack of centrioles in this Drosophila cell line. Recent advances, such as the characterization of the mh gene and the genomic analysis of 1182-4 cells, allow now a better understanding of the physiology of these cells. By combining these new data, we propose three reasonable hypotheses of the genesis of this remarkable phenotype.

ProDCoNN: Protein design using a convolutional neural network.

Designing protein sequences that fold to a given three-dimensional (3D) structure has long been a challenging problem in computational structural biology with significant theoretical and practical implications. In this study, we first formulated this problem as predicting the residue type given the 3D structural environment around the C α atom of a residue, which is repeated for each residue of a protein. We designed a nine-layer 3D deep convolutional neural network (CNN) that takes as input a gridded box with the atomic coordinates and types around a residue. Several CNN layers were designed to capture structure information at different scales, such as bond lengths, bond angles, torsion angles, and secondary structures. Trained on a very large number of protein structures, the method, called ProDCoNN (protein design with CNN), achieved state-of-the-art performance when tested on large numbers of test proteins and benchmark datasets.

Segmentation by classification: A novel and reliable approach for semi-automatic selection of HIV/SIV envelope spikes.

We describe a semiautomated approach to segment Env spikes from the membrane envelope of Simian Immunodeficiency Virus visualized by cryoelectron tomography of frozen-hydrated specimens. Multivariate data analysis is applied to a large set of overlapping subvolumes extracted semiautomatically from the viral envelope and does not utilize a template of the target structure. The major manual step used in the method involves determination of six points that define an ellipsoid approximating the virion shape. The approach is robust to departures of the actual virion from this starting ellipsoid. A point cage of sufficient density is generated to ensure that any spike-like protein is identified multiple times. Subsequently translational alignment of class averages to a cylindrical reference on a curved surface separates subvolumes with spikes from those without. Spike containing subvolumes identified multiple times are removed by proximity analysis. Slightly different procedures segment spikes in the equatorial and the polar regions. Once all spikes are segmented, further alignment of class averages using separately the polar and spin angles produces recognizable spike images. Our approach localized 96% of the equatorial spikes and 85% of all spikes identified manually; it identifies a significant number of additional spikes missed by manual selection. Two types of spike shapes were segmented, one with near 3-fold symmetry resembling the conventional spike, the other had a T-shape resembling the spike structure obtained when antibodies such as PG9 bind to HIV Env. The approach should be applicable to segmentation of any protein spikes extending from a cellular or virion envelope.

Biomedical word sense disambiguation with bidirectional long short-term memory and attention-based neural networks.

BACKGROUND: In recent years, deep learning methods have been applied to many natural language processing tasks to achieve state-of-the-art performance. However, in the biomedical domain, they have not out-performed supervised word sense disambiguation (WSD) methods based on support vector machines or random forests, possibly due to inherent similarities of medical word senses. RESULTS: In this paper, we propose two deep-learning-based models for supervised WSD: a model based on bi-directional long short-term memory (BiLSTM) network, and an attention model based on self-attention architecture. Our result shows that the BiLSTM neural network model with a suitable upper layer structure performs even better than the existing state-of-the-art models on the MSH WSD dataset, while our attention model was 3 or 4 times faster than our BiLSTM model with good accuracy. In addition, we trained "universal" models in order to disambiguate all ambiguous words together. That is, we concatenate the embedding of the target ambiguous word to the max-pooled vector in the universal models, acting as a "hint". The result shows that our universal BiLSTM neural network model yielded about 90 percent accuracy. CONCLUSION: Deep contextual models based on sequential information processing methods are able to capture the relative contextual information from pre-trained input word embeddings, in order to provide state-of-the-art results for supervised biomedical WSD tasks.

Adhesive Cements That Bond Soft Tissue Ex Vivo.

The aim of the present study was to evaluate the soft tissue bond strength of a newly developed, monomeric, biomimetic, tissue adhesive called phosphoserine modified cement (PMC). Two types of PMCs were evaluated using lap shear strength (LSS) testing, on porcine skin: a calcium metasilicate (CS1), and alpha tricalcium phosphate (αTCP) PMC. CS1 PCM bonded strongly to skin, reaching a peak LSS of 84, 132, and 154 KPa after curing for 0.5, 1.5, and 4 h, respectively. Cyanoacrylate and fibrin glues reached an LSS of 207 kPa and 33 kPa, respectively. αTCP PMCs reached a final LSS of ≈110 kPa. In soft tissues, stronger bond strengths were obtained with αTCP PMCs containing large amounts of amino acid (70-90 mol%), in contrast to prior studies in calcified tissues (30-50 mol%). When αTCP particle size was reduced by wet milling, and for CS1 PMCs, the strongest bonding was obtained with mole ratios of 30-50% phosphoserine. While PM-CPCs behave like stiff ceramics after setting, they bond to soft tissues, and warrant further investigation as tissue adhesives, particularly at the interface between hard and soft tissues.

Evaluating semantic relations in neural word embeddings with biomedical and general domain knowledge bases.

BACKGROUND: In the past few years, neural word embeddings have been widely used in text mining. However, the vector representations of word embeddings mostly act as a black box in downstream applications using them, thereby limiting their interpretability. Even though word embeddings are able to capture semantic regularities in free text documents, it is not clear how different kinds of semantic relations are represented by word embeddings and how semantically-related terms can be retrieved from word embeddings. METHODS: To improve the transparency of word embeddings and the interpretability of the applications using them, in this study, we propose a novel approach for evaluating the semantic relations in word embeddings using external knowledge bases: Wikipedia, WordNet and Unified Medical Language System (UMLS). We trained multiple word embeddings using health-related articles in Wikipedia and then evaluated their performance in the analogy and semantic relation term retrieval tasks. We also assessed if the evaluation results depend on the domain of the textual corpora by comparing the embeddings of health-related Wikipedia articles with those of general Wikipedia articles. RESULTS: Regarding the retrieval of semantic relations, we were able to retrieve diverse semantic relations in the nearest neighbors of a given word. Meanwhile, the two popular word embedding approaches, Word2vec and GloVe, obtained comparable results on both the analogy retrieval task and the semantic relation retrieval task, while dependency-based word embeddings had much worse performance in both tasks. We also found that the word embeddings trained with health-related Wikipedia articles obtained better performance in the health-related relation retrieval tasks than those trained with general Wikipedia articles. CONCLUSION: It is evident from this study that word embeddings can group terms with diverse semantic relations together. The domain of the training corpus does have impact on the semantic relations represented by word embeddings. We thus recommend using domain-specific corpus to train word embeddings for domain-specific text mining tasks.

Correction to: Evaluating semantic relations in neural word embeddings with biomedical and general domain knowledge bases.

After publication of this supplement article [1], it was brought to our attention that the Results section of the abstract contained a partial sentence.

Inhibition of lysine-specific demethylase 1 prevents proliferation and mediates cisplatin sensitivity in ovarian cancer cells.

Lysine-specific demethylase 1 (LSD1) functions as a transcriptional coregulator by modulating histone methylation and has been associated with numerous high-risk cancers. Previously, our group and others identified LSD1 as an upregulated gene in ovarian cancer, and reported that the upregulation of LSD1 was associated with poor prognosis of patients with ovarian cancer. However, the role of LSD1 in ovarian cancer requires further investigation. The present study revealed that the overexpression of LSD1 significantly promoted the proliferation of SKOV3 ovarian cancer cells, while knockdown of LSD1 markedly inhibited cell proliferation and potentiated cisplatin-induced cell apoptosis, supporting LSD1 as an oncogenic protein in ovarian cancer. Mechanistic studies have indicated that LSD1 modulates the expression of cyclin dependent kinase inhibitor 1, Survivin, B-cell lymphoma-2 (Bcl-2) and Bcl-2-associated X genes, which are known regulators of cell proliferation. Furthermore, LSD1 knockdown plus cisplatin synergistically impaired cell migration via the induction of the epithelial marker E-cadherin and inhibition of the mesenchymal markers, snail family transcriptional repressor 1 and Vimentin. These data of the present study indicated LSD1 as a potential regulator of ovarian cancer cell progression and suggested an unfavorable role of LSD1 in cisplatin-based regimens.

Land cover classification from multi-temporal, multi-spectral remotely sensed imagery using patch-based recurrent neural networks.

Environmental sustainability research is dependent on accurate land cover information. Even with the increased number of satellite systems and sensors acquiring data with improved spectral, spatial, radiometric and temporal characteristics and the new data distribution policy, most existing land cover datasets are derived from a pixel-based, single-date multi-spectral remotely sensed image with an unacceptable accuracy. One major bottleneck for accuracy improvement is how to develop an accurate and effective image classification protocol. By incorporating and utilizing multi-spectral, multi-temporal and spatial information in remote sensing images and considering the inherit spatial and sequential interdependence among neighboring pixels, we propose a new patch-based recurrent neural network (PB-RNN) system tailored for classifying multi-temporal remote sensing data. The system is designed by incorporating distinctive characteristics of multi-temporal remote sensing data. In particular, it uses multi-temporal-spectral-spatial samples and deals with pixels contaminated by clouds/shadow present in multi-temporal data series. Using a Florida Everglades ecosystem study site covering an area of 771 square kilometers, the proposed PB-RNN system has achieved a significant improvement in the classification accuracy over a pixel-based recurrent neural network (RNN) system, a pixel-based single-image neural network (NN) system, a pixel-based multi-image NN system, a patch-based single-image NN system, and a patch-based multi-image NN system. For example, the proposed system achieves 97.21% classification accuracy while the pixel-based single-image NN system achieves 64.74%. By utilizing methods like the proposed PB-RNN one, we believe that much more accurate land cover datasets can be produced over large areas.

A patch-based convolutional neural network for remote sensing image classification.

Availability of accurate land cover information over large areas is essential to the global environment sustainability; digital classification using medium-resolution remote sensing data would provide an effective method to generate the required land cover information. However, low accuracy of existing per-pixel based classification methods for medium-resolution data is a fundamental limiting factor. While convolutional neural networks (CNNs) with deep layers have achieved unprecedented improvements in object recognition applications that rely on fine image structures, they cannot be applied directly to medium-resolution data due to lack of such fine structures. In this paper, considering the spatial relation of a pixel to its neighborhood, we propose a new deep patch-based CNN system tailored for medium-resolution remote sensing data. The system is designed by incorporating distinctive characteristics of medium-resolution data; in particular, the system computes patch-based samples from multidimensional top of atmosphere reflectance data. With a test site from the Florida Everglades area (with a size of 771 square kilometers), the proposed new system has outperformed pixel-based neural network, pixel-based CNN and patch-based neural network by 24.36%, 24.23% and 11.52%, respectively, in overall classification accuracy. By combining the proposed deep CNN and the huge collection of medium-resolution remote sensing data, we believe that much more accurate land cover datasets can be produced over large areas.

A harmonized segmentation protocol for hippocampal and parahippocampal subregions: Why do we need one and what are the key goals?

The advent of high-resolution magnetic resonance imaging (MRI) has enabled in vivo research in a variety of populations and diseases on the structure and function of hippocampal subfields and subdivisions of the parahippocampal gyrus. Because of the many extant and highly discrepant segmentation protocols, comparing results across studies is difficult. To overcome this barrier, the Hippocampal Subfields Group was formed as an international collaboration with the aim of developing a harmonized protocol for manual segmentation of hippocampal and parahippocampal subregions on high-resolution MRI. In this commentary we discuss the goals for this protocol and the associated key challenges involved in its development. These include differences among existing anatomical reference materials, striking the right balance between reliability of measurements and anatomical validity, and the development of a versatile protocol that can be adopted for the study of populations varying in age and health. The commentary outlines these key challenges, as well as the proposed solution of each, with concrete examples from our working plan. Finally, with two examples, we illustrate how the harmonized protocol, once completed, is expected to impact the field by producing measurements that are quantitatively comparable across labs and by facilitating the synthesis of findings across different studies. © 2016 Wiley Periodicals, Inc.

LSD1-mediated epigenetic modification contributes to ovarian cancer cell migration and invasion.

Lysine-specific demethylase 1 (LSD1) has been implicated in the process of tumor progression at various steps, but its role in epithelial-messenchymal transition (EMT) and the migration of ovarian cancer cells remains obscure. In this study, we demonstrated the effect of LSD1 on ovarian cancer cell migration and the regulatory role of LSD1 in the expression of EMT markers. Inhibition of LSD1 expression impaired the migration and invasion of HO8910 ovarian cancer cells. In contrast, overexpression of LSD1 enhanced the cell migration and invasion of HO8910 cells. Mechanistic analyses showed that LSD1 promoted cell migration through induction of N-cadherin, vimentin, MMP-2 and inhibition of E-cadherin. Furthermore, LSD1 interacted with the promoter of E-cadherin and demethylated histone H3 lysine 4 (H3K4) at this region, downregulated E-cadherin expression, and consequently enhanced ovarian cancer cell migration. These data indicate that LSD1 acts as an epigenetic regulator of EMT and contributes to the metastasis of ovarian cancer.