Single-Walled Cluster Nanotubes for Single-Atom Catalysts with Precise Structures.

Spatially confining isolated atomic sites in low-dimensional nanostructures is a promising strategy for preparing high-performance single-atom catalysts (SACs). Herein, fascinating polyoxometalate cluster-based single-walled nanotubes (POM-SWNTs) with atomically precise structures, uniform diameter, and single-cluster wall thickness are constructed by lacunary POM clusters (PW11 and P2W17 clusters). Isolated metal centers are accurately incorporated into the PW11-SWNTs and P2W17-SWNTs supports. The structures of the resulting MPW11-SWNTs and MP2W17-SWNTs are well established (M = Cu, Pt). Molecular dynamics simulations demonstrate the stability of POM-SWNTs. Furthermore, the turnover frequency of PtP2W17-SWNTs is 20 times higher than that of PtP2W17 cluster units and 140 times higher than that of Pt nanoparticles in the alcoholysis of dimethylphenylsilane. Theoretical studies indicate that incorporating a Pt atom into the P2W17 support induces straightforward electron transfer between them, combining the nanoconfined environment to enhance the catalytic activity of PtP2W17-SWNTs. This work shows the feasibility of using subnanometric POM clusters to assemble single-walled cluster nanotubes, highlighting their potential to prepare superior SACs with precise structures.

Single Cell Self-Paced Clustering with Transcriptome Sequencing Data.

Single cell RNA sequencing (scRNA-seq) allows researchers to explore tissue heterogeneity, distinguish unusual cell identities, and find novel cellular subtypes by providing transcriptome profiling for individual cells. Clustering analysis is usually used to predict cell class assignments and infer cell identities. However, the performance of existing single-cell clustering methods is extremely sensitive to the presence of noise data and outliers. Existing clustering algorithms can easily fall into local optimal solutions. There is still no consensus on the best performing method. To address this issue, we introduce a single cell self-paced clustering (scSPaC) method with F-norm based nonnegative matrix factorization (NMF) for scRNA-seq data and a sparse single cell self-paced clustering (sscSPaC) method with l21-norm based nonnegative matrix factorization for scRNA-seq data. We gradually add single cells from simple to complex to our model until all cells are selected. In this way, the influences of noisy data and outliers can be significantly reduced. The proposed method achieved the best performance on both simulation data and real scRNA-seq data. A case study about human clara cells and ependymal cells scRNA-seq data clustering shows that scSPaC is more advantageous near the clustering dividing line.

Endogenous miRNA-Activated DNA Nanomachine for Intracellular miRNA Imaging and Gene Silencing.

The development of multifunctional nanoplatforms that integrate both diagnostic and therapeutic functions has always been extremely desirable and challenging in the cancer combat. Here, we report an endogenous miRNA-activated DNA nanomachine (EMDN) in living cells for concurrent sensitive miRNA imaging and activatable gene silencing. EMDN is constructed by interval hybridization of two functional DNA monomers (R/HP and F) to a DNA nanowire generated by hybridization chain reaction. After the target cell-specific transportation of EMDN, intracellular let-7a miRNA initiates the DNA nanomachine by DNA strand displacement cascades, resulting in an amplified fluorescence resonance energy-transfer signal and the release of many free HP sequences. The restoration of HP hairpin structures further activates the split-DNAzyme to identify and cleave the EGR-1 mRNA to realize gene silencing therapy. The proposed EMDN shows efficient cell internalization, good biological stability, rapid reaction kinetics, and the ability to avoid false-positive signals, thus ensuring reliable miRNA imaging in living cells. Meanwhile, the controlled activation of the split-DNAzyme activity regulated by the intracellular specific miRNA may be promising in the precise treatment of cancer. Collectively, this strategy provides a valuable nanoplatform for early clinical diagnosis and activatable gene therapy of tumors.

Identification of a New DNA Aptamer by Tissue-SELEX for Cancer Recognition and Imaging.

Cancer has become one of the most common diseases with high mortality in humans. Early and accurate diagnosis of cancer is of great significance to enhance the survival rate of patients. Therefore, effective molecular ligands capable of selectively recognizing cancer are urgently needed. In this work, we identified a new DNA aptamer named SW1 by tissue-based systematic evolution of ligands by exponential enrichment (tissue-SELEX), in which cancerous liver tissue sections were used as the positive control and adjacent normal liver tissue sections were used as the negative control. Taking immobilized liver cancer SMMC-7721 cells as the research object, aptamer SW1 exhibited excellent affinity with a Kd value of 123.62 ± 17.53 nM, and its binding target was preliminarily determined as a non-nucleic acid substance in the nucleus. Moreover, tissue imaging results showed that SW1 explicitly recognized cancerous liver tissues with a high detection rate of 72.7% but displayed a low detection rate to adjacent normal tissues. In addition to liver cancer cells and tissues, aptamer SW1 has been demonstrated to recognize various other types of cancer cells and tissues. Furthermore, SW1-A, an optimized aptamer of SW1, maintained its excellent affinity toward liver cancer cells and tissues. Collectively, these results indicate that SW1 possesses great potential for use as an effective molecular probe for clinical diagnosis of cancer.

Direct Ink Writing of Li1.3 Al0.3 Ti1.7 (PO4 )3 -Based Solid-State Electrolytes with Customized Shapes and Remarkable Electrochemical Behaviors.

All-solid-state lithium batteries have received extensive attention due to their high safety and promising energy density and are considered as the next-generation electrochemical energy storage system. However, exploring solid-state electrolytes in customized geometries without sacrificing the ionic transport is significant yet challenging. Herein, various 3D printable Li1.3 Al0.3 Ti1.7 (PO4 )3 (LATP)-based inks are developed to construct ceramic and hybrid solid-state electrolytes with arbitrary shapes as well as high conductivities. The obtained inks show suitable rheological behaviors and can be successfully extruded into solid-state electrolytes using the direct ink writing (DIW) method. As-printed free-standing LATP ceramic solid-state electrolytes deliver high ionic conductivity up to 4.24 × 10-4  S cm-1 and different shapes such as "L", "T," and "+" can be easily realized without sacrificing high ionic transport properties. Moreover, using this printing method, LATP-based hybrid solid-state electrolytes can be directly printed on LiFePO4 cathodes for solid-state lithium batteries, where a high discharge capacity of 150 mAh g-1 at 0.5 C is obtained. The DIW strategy for solid-state electrolytes demonstrates a new way toward advanced solid-state energy storage with the high ionic transport and customized manufacturing ability.

TrioMDR: Detecting SNP interactions in trio families with model-based multifactor dimensionality reduction.

Single nucleotide polymorphism (SNP) interactions can explain the missing heritability of common complex diseases. Many interaction detection methods have been proposed in genome-wide association studies, and they can be divided into two types: population-based and family-based. Compared with population-based methods, family-based methods are robust vs. population stratification. Several family-based methods have been proposed, among which Multifactor Dimensionality Reduction (MDR)-based methods are popular and powerful. However, current MDR-based methods suffer from heavy computational burden. Furthermore, they do not allow for main effect adjustment. In this work we develop a two-stage model-based MDR approach (TrioMDR) to detect multi-locus interaction in trio families (i.e., two parents and one affected child). TrioMDR combines the MDR framework with logistic regression models to check interactions, so TrioMDR can adjust main effects. In addition, unlike consuming permutation procedures used in traditional MDR-based methods, TrioMDR utilizes a simple semi-parameter P-values correction procedure to control type I error rate, this procedure only uses a few permutations to achieve the significance of a multi-locus model and significantly speeds up TrioMDR. We performed extensive experiments on simulated data to compare the type I error and power of TrioMDR under different scenarios. The results demonstrate that TrioMDR is fast and more powerful in general than some recently proposed methods for interaction detection in trios. The R codes of TrioMDR are available at: https://github.com/TrioMDR/TrioMDR.

Drug repositioning based on individual bi-random walks on a heterogeneous network.

BACKGROUND: Traditional drug research and development is high cost, time-consuming and risky. Computationally identifying new indications for existing drugs, referred as drug repositioning, greatly reduces the cost and attracts ever-increasing research interests. Many network-based methods have been proposed for drug repositioning and most of them apply random walk on a heterogeneous network consisted with disease and drug nodes. However, these methods generally adopt the same walk-length for all nodes, and ignore the different contributions of different nodes. RESULTS: In this study, we propose a drug repositioning approach based on individual bi-random walks (DR-IBRW) on the heterogeneous network. DR-IBRW firstly quantifies the individual work-length of random walks for each node based on the network topology and knowledge that similar drugs tend to be associated with similar diseases. To account for the inner structural difference of the heterogeneous network, it performs bi-random walks with the quantified walk-lengths, and thus to identify new indications for approved drugs. Empirical study on public datasets shows that DR-IBRW achieves a much better drug repositioning performance than other related competitive methods. CONCLUSIONS: Using individual random walk-lengths for different nodes of heterogeneous network indeed boosts the repositioning performance. DR-IBRW can be easily generalized to prioritize links between nodes of a network.

Protein-Protein Interactions Prediction Using a Novel Local Conjoint Triad Descriptor of Amino Acid Sequences.

Protein-protein interactions (PPIs) play crucial roles in almost all cellular processes. Although a large amount of PPIs have been verified by high-throughput techniques in the past decades, currently known PPIs pairs are still far from complete. Furthermore, the wet-lab experiments based techniques for detecting PPIs are time-consuming and expensive. Hence, it is urgent and essential to develop automatic computational methods to efficiently and accurately predict PPIs. In this paper, a sequence-based approach called DNN-LCTD is developed by combining deep neural networks (DNNs) and a novel local conjoint triad description (LCTD) feature representation. LCTD incorporates the advantage of local description and conjoint triad, thus, it is capable to account for the interactions between residues in both continuous and discontinuous regions of amino acid sequences. DNNs can not only learn suitable features from the data by themselves, but also learn and discover hierarchical representations of data. When performing on the PPIs data of Saccharomyces cerevisiae, DNN-LCTD achieves superior performance with accuracy as 93.12%, precision as 93.75%, sensitivity as 93.83%, area under the receiver operating characteristic curve (AUC) as 97.92%, and it only needs 718 s. These results indicate DNN-LCTD is very promising for predicting PPIs. DNN-LCTD can be a useful supplementary tool for future proteomics study.

Deep Clustering: A Comprehensive Survey.

Cluster analysis plays an indispensable role in machine learning and data mining. Learning a good data representation is crucial for clustering algorithms. Recently, deep clustering (DC), which can learn clustering-friendly representations using deep neural networks (DNNs), has been broadly applied in a wide range of clustering tasks. Existing surveys for DC mainly focus on the single-view fields and the network architectures, ignoring the complex application scenarios of clustering. To address this issue, in this article, we provide a comprehensive survey for DC in views of data sources. With different data sources, we systematically distinguish the clustering methods in terms of methodology, prior knowledge, and architecture. Concretely, DC methods are introduced according to four categories, i.e., traditional single-view DC, semi-supervised DC, deep multiview clustering (MVC), and deep transfer clustering. Finally, we discuss the open challenges and potential future opportunities in different fields of DC.

Peripheral and tumor-infiltrating immune cells are correlated with patient outcomes in ovarian cancer.

OBJECTIVE: At present, there is still a lack of reliable biomarkers for ovarian cancer (OC) to guide prognosis prediction and accurately evaluate the dominant population of immunotherapy. In recent years, the relationship between peripheral blood markers and tumor-infiltrating immune cells (TICs) with cancer has attracted much attention. However, the relationship between the survival of OC patients and intratumoral- or extratumoral-associated immune cells remains controversial. METHODS: In this study, four machine-learning algorithms were used to predict overall survival in OC patients based on peripheral blood indicators. To further screen out immune-related gene and molecular targets, we systematically explored the correlation between TICs and OC patient survival based on The Cancer Genome Atlas database. Using the TICs score method, patients were divided into a low immune infiltrating cell group and a high immune infiltrating cell group. RESULTS: The results showed that there was a significant statistical significance between the peripheral blood indicators and the survival prognosis of OC patients. Survival analysis showed that TICs play a crucial role in the survival of OC patients. Four core genes, CXCL9, CD79A, MS4A1, and MZB1, were identified by cross-PPI and COX regression analysis. Further analysis found that these genes were significantly associated with both TICs and survival in OC patients. CONCLUSIONS: These results suggest that both peripheral blood markers and TICs can be used as prognostic predictors in patients with OC, and CXCL9, CD79A, MS4A1, and MZB1 may be potential therapeutic targets for OC immunotherapy.

Boosting adversarial robustness via self-paced adversarial training.

Adversarial training is considered one of the most effective methods to improve the adversarial robustness of deep neural networks. Despite the success, it still suffers from unsatisfactory performance and overfitting. Considering the intrinsic mechanism of adversarial training, recent studies adopt the idea of curriculum learning to alleviate overfitting. However, this also introduces new issues, that is, lacking the quantitative criterion for attacks' strength and catastrophic forgetting. To mitigate such issues, we propose the self-paced adversarial training (SPAT), which explicitly builds the learning process of adversarial training based on adversarial examples of the whole dataset. Specifically, our model is first trained with "easy" adversarial examples, and then is continuously enhanced by gradually adding "complex" adversarial examples. This way strengthens the ability to fit "complex" adversarial examples while holding in mind "easy" adversarial samples. To balance adversarial examples between classes, we determine the difficulty of the adversarial examples locally in each class. Notably, this learning paradigm can also be incorporated into other advanced methods for further boosting adversarial robustness. Experimental results show the effectiveness of our proposed model against various attacks on widely-used benchmarks. Especially, on CIFAR100, SPAT provides a boost of 1.7% (relatively 5.4%) in robust accuracy on the PGD10 attack and 3.9% (relatively 7.2%) in natural accuracy for AWP.

Research progress of quercetin in cardiovascular disease.

Quercetin is one of the most common flavonoids. More and more studies have found that quercetin has great potential utilization value in cardiovascular diseases (CVD), such as antioxidant, antiplatelet aggregation, antibacterial, cholesterol lowering, endothelial cell protection, etc. However, the medicinal value of quercetin is mostly limited to animal models and preclinical studies. Due to the complexity of the human body and functional structure compared to animals, more research is needed to explore whether quercetin has the same mechanism of action and pharmacological value as animal experiments. In order to systematically understand the clinical application value of quercetin, this article reviews the research progress of quercetin in CVD, including preclinical and clinical studies. We will focus on the relationship between quercetin and common CVD, such as atherosclerosis, myocardial infarction, ischemia reperfusion injury, heart failure, hypertension and arrhythmia, etc. By elaborating on the pathophysiological mechanism and clinical application research progress of quercetin's protective effect on CVD, data support is provided for the transformation of quercetin from laboratory to clinical application.

CDDnet: Cross-domain denoising network for low-dose CT image via local and global information alignment.

The domain shift problem has emerged as a challenge in cross-domain low-dose CT (LDCT) image denoising task, where the acquisition of a sufficient number of medical images from multiple sources may be constrained by privacy concerns. In this study, we propose a novel cross-domain denoising network (CDDnet) that incorporates both local and global information of CT images. To address the local component, a local information alignment module has been proposed to regularize the similarity between extracted target and source features from selected patches. To align the general information of the semantic structure from a global perspective, an autoencoder is adopted to learn the latent correlation between the source label and the estimated target label generated by the pre-trained denoiser. Experimental results demonstrate that our proposed CDDnet effectively alleviates the domain shift problem, outperforming other deep learning-based and domain adaptation-based methods under cross-domain scenarios.

Adaptive Feature Projection with Distribution Alignment for Deep Incomplete Multi-view Clustering.

Incomplete multi-view clustering (IMVC) analysis, where some views of multi-view data usually have missing data, has attracted increasing attention. However, existing IMVC methods still have two issues: (1) they pay much attention to imputing or recovering the missing data, without considering the fact that the imputed values might be inaccurate due to the unknown label information, (2) the common features of multiple views are always learned from the complete data, while ignoring the feature distribution discrepancy between the complete and incomplete data. To address these issues, we propose an imputation-free deep IMVC method and consider distribution alignment in feature learning. Concretely, the proposed method learns the features for each view by autoencoders and utilizes an adaptive feature projection to avoid the imputation for missing data. All available data are projected into a common feature space, where the common cluster information is explored by maximizing mutual information and the distribution alignment is achieved by minimizing mean discrepancy. Additionally, we design a new mean discrepancy loss for incomplete multi-view learning and make it applicable in mini-batch optimization. Extensive experiments demonstrate that our method achieves the comparable or superior performance compared with state-of-the-art methods.

Automatic tumor segmentation and metachronous single-organ metastasis prediction of nasopharyngeal carcinoma patients based on multi-sequence magnetic resonance imaging.

Background: Distant metastases is the main failure mode of nasopharyngeal carcinoma. However, early prediction of distant metastases in NPC is extremely challenging. Deep learning has made great progress in recent years. Relying on the rich data features of radiomics and the advantages of deep learning in image representation and intelligent learning, this study intends to explore and construct the metachronous single-organ metastases (MSOM) based on multimodal magnetic resonance imaging. Patients and methods: The magnetic resonance imaging data of 186 patients with nasopharyngeal carcinoma before treatment were collected, and the gross tumor volume (GTV) and metastatic lymph nodes (GTVln) prior to treatment were defined on T1WI, T2WI, and CE-T1WI. After image normalization, the deep learning platform Python (version 3.9.12) was used in Ubuntu 20.04.1 LTS to construct automatic tumor detection and the MSOM prediction model. Results: There were 85 of 186 patients who had MSOM (including 32 liver metastases, 25 lung metastases, and 28 bone metastases). The median time to MSOM was 13 months after treatment (7-36 months). The patients were randomly assigned to the training set (N = 140) and validation set (N = 46). By comparison, we found that the overall performance of the automatic tumor detection model based on CE-T1WI was the best (6). The performance of automatic detection for primary tumor (GTV) and lymph node gross tumor volume (GTVln) based on the CE-T1WI model was better than that of models based on T1WI and T2WI (AP@0.5 is 59.6 and 55.6). The prediction model based on CE-T1WI for MSOM prediction achieved the best overall performance, and it obtained the largest AUC value (AUC = 0.733) in the validation set. The precision, recall, precision, and AUC of the prediction model based on CE-T1WI are 0.727, 0.533, 0.730, and 0.733 (95% CI 0.557-0.909), respectively. When clinical data were added to the deep learning prediction model, a better performance of the model could be obtained; the AUC of the integrated model based on T2WI, T1WI, and CE-T1WI were 0.719, 0.738, and 0.775, respectively. By comparing the 3-year survival of high-risk and low-risk patients based on the fusion model, we found that the 3-year DMFS of low and high MSOM risk patients were 95% and 11.4%, respectively (p < 0.001). Conclusion: The intelligent prediction model based on magnetic resonance imaging alone or combined with clinical data achieves excellent performance in automatic tumor detection and MSOM prediction for NPC patients and is worthy of clinical application.

EEGDnet: Fusing non-local and local self-similarity for EEG signal denoising with transformer.

Electroencephalogram (EEG) has shown a useful approach to produce a brain-computer interface (BCI). One-dimensional (1-D) EEG signal is yet easily disturbed by certain artifacts (a.k.a. noise) due to the high temporal resolution. Thus, it is crucial to remove the noise in received EEG signal. Recently, deep learning-based EEG signal denoising approaches have achieved impressive performance compared with traditional ones. It is well known that the characteristics of self-similarity (including non-local and local ones) of data (e.g., natural images and time-domain signals) are widely leveraged for denoising. However, existing deep learning-based EEG signal denoising methods ignore either the non-local self-similarity (e.g., 1-D convolutional neural network) or local one (e.g., fully connected network and recurrent neural network). To address this issue, we propose a novel 1-D EEG signal denoising network with 2-D transformer, namely EEGDnet. Specifically, we comprehensively take into account the non-local and local self-similarity of EEG signal through the transformer module. By fusing non-local self-similarity in self-attention blocks and local self-similarity in feed forward blocks, the negative impact caused by noises and outliers can be reduced significantly. Extensive experiments show that, compared with other state-of-the-art models, EEGDnet achieves much better performance in terms of both quantitative and qualitative metrics. Specifically, EEGDnet can achieve 18% and 11% improvements in correlation coefficients when removing ocular artifacts and muscle artifacts, respectively.

BS-80K: The first large open-access dataset of bone scan images.

BACKGROUND: Radionuclide bone scanning is one of the most common tools in the inspection of bone metastasis. Conventionally, the analysis of bone scan image is derived from manual diagnosing. However, this task requires extensive subjective diagnostic experience and is extremely time-consuming. To this end, a series of studies concerning computer-aided diagnosis via machine learning tools have been proposed. Although some inspiring progress has been achieved, the implemented bone scan image datasets in these research areas are generally too small, private or non-general, which limits their practical significance and impedes the follow-up research. METHOD: To address this issue, we present a large, publicly available and general dataset consisting of 82544 bone scan images associated with 3247 patients from West China Hospital, named BS-80K. In BS-80K, each patient provides two whole bone scan images corresponding to the anterior view (ANT) and the posterior view (POST). For each view, there are 13 region-wise slices of the body parts susceptible to bone metastasis. Based on an authorized original labeling criterion, labels annotated by experienced specialists are offered with the images. Moreover, within each whole body image, multiple bounding boxes containing suspectable hot spots and their annotations are supplied as well. All images in BS-80K have been de-identified to protect patients' privacy. RESULTS: Based on 6 popular deep learning models for classification and object detection, we provide the benchmark for a number of computer-aided medical tasks, including general bone metastasis prediction and object detection for whole body images, and specific bone metastasis prediction for different body parts. According to extensive experiments, the adopted classification models achieve remarkable results in accuracy and specificity (around 95%) on most metastasis prediction tasks, which are approximate to the average ability of corresponding specialists. As for the object detection task, the best average precision of the adopted models reaches 0.2484 and the lowest is 0.1334. DISCUSSION: Through the comparison of metastasis prediction performance between the benchmark and related work, we observe that the widely used models trained by BS-80K achieve significantly better results than the elaborately designed models trained by smaller datasets. This indicates that with the large amount of data, BS-80K has great potential to galvanize the research about computer-aided analysis on bone scan image. CONCLUSION: To the best of our knowledge, BS-80K is the first large publicly available dataset of bone scanning, which favors a wide range of research on computer-aided bone metastasis diagnosis. The full dataset is now available at https://drive.google.com/drive/folders/1DOBkLXgQeREQjF-nQIGNBBzPCb5s7RNu?usp=sharing.

Dual self-paced multi-view clustering.

By utilizing the complementary information from multiple views, multi-view clustering (MVC) algorithms typically achieve much better clustering performance than conventional single-view methods. Although in this field, great progresses have been made in past few years, most existing multi-view clustering methods still suffer the following shortcomings: (1) most MVC methods are non-convex and thus are easily stuck into suboptimal local minima; (2) the effectiveness of these methods is sensitive to the existence of noises or outliers; and (3) the qualities of different features and views are usually ignored, which can also influence the clustering result. To address these issues, we propose dual self-paced multi-view clustering (DSMVC) in this paper. Specifically, DSMVC takes advantage of self-paced learning to tackle the non-convex issue. By applying a soft-weighting scheme of self-paced learning for instances, the negative impact caused by noises and outliers can be significantly reduced. Moreover, to alleviate the feature and view quality issues, we develop a novel feature selection approach in a self-paced manner and a weighting term for views. Experimental results on real-world data sets demonstrate the effectiveness of the proposed method.

Highly sensitive detection of cancer cells via split aptamer mediated proximity-induced hybridization chain reaction.

Highly sensitive detection of cancer cells is of great importance for evaluating cancer development and improving survival rates. Here, we developed a split aptamer mediated proximity-induced hybridization chain reaction (HCR) strategy to meet this purpose. In this strategy, two split aptamer initiator probes, Sp-a and Sp-b, and two HCR hairpin probes, H1 and H2 were designed. The split aptamer initiator probes contained two components, split aptamer domains being responsible for target recognition, and the split initiator parts serving as the HCR promoter. In the presence of target cells, Sp-a and Sp-b would self-assemble on the cell surfaces, allowing the formation of an intact nicked initiator to activate the HCR reaction. Benefit from low background split aptamers and HCR amplification, this strategy presented high sensitivity in quantitative detection with a detection limit of 18 cells in 150 µL of binding buffer. Moreover, the approach exhibited excellent specificity to target cells in 10% fetal bovine serum and mixed cell samples, which was favorable for clinical diagnosis in complex biological environment. In addition, by changing the split aptamers attached to the split initiator, the proposed strategy can be expanded to detect various kinds of target cells. It may provide a novel and useful applicable platform for the sensitive detection of cancer cells in biomedicine and tumor-related studies.

Machine learning based early mortality prediction in the emergency department.

BACKGROUND: It is a great challenge for emergency physicians to early detect the patient's deterioration and prevent unexpected death through a large amount of clinical data, which requires sufficient experience and keen insight. OBJECTIVE: To evaluate the performance of machine learning models in quantifying the severity of emergency department (ED) patients and identifying high-risk patients. METHODS: Using routinely-available demographics, vital signs and laboratory tests extracted from electronic health records (EHRs), a framework based on machine learning and feature engineering was proposed for mortality prediction. Patients who had one complete record of vital signs and laboratory tests in ED were included. The following patients were excluded: pediatric patients aged < 18 years, pregnant woman, and patients died or were discharged or hospitalized within 12 h after admission. Based on 76 original features extracted, 9 machine learning models were adopted to validate our proposed framework. Their optimal hyper-parameters were fine-tuned using the grid search method. The prediction results were evaluated on performance metrics (i.e., accuracy, area under the curve (AUC), recall and precision) with repeated 5-fold cross-validation (CV). The time window from patient admission to the prediction was analyzed at 12 h, 24 h, 48 h, and entire stay. RESULTS: We studied a total of 1114 ED patients with 71.54% (797/1114) survival and 28.46% (317/1114) death in the hospital. The results revealed a more complete time window leads to better prediction performance. Using the entire stay records, the LightGBM model with refined feature engineering demonstrated high discrimination and achieved 93.6% (±0.008) accuracy, 97.6% (±0.003) AUC, 97.1% (±0.008) recall, and 94.2% (±0.006) precision, even if no diagnostic information was utilized. CONCLUSIONS: This study quantifies the criticality of ED patients and appears to have significant potential as a clinical decision support tool in assisting physicians in their clinical routine. While the model requires validation before use elsewhere, the same methodology could be used to create a strong model for the new hospital.