arXiv subjects

Xian Yang

Publications and source records attributed to Xian Yang.

49 records · Page 3Linked to original sources

arXiv2022

Label Dependent Attention Model for Disease Risk Prediction Using Multimodal Electronic Health Records

Disease risk prediction has attracted increasing attention in the field of modern healthcare, especially with the latest advances in artificial intelligence (AI). Electronic health records (EHRs), which contain heterogeneous patient information, are widely used in disease risk prediction tasks. One challenge of applying AI models for risk prediction lies in generating interpretable evidence to support the prediction results while retaining the prediction ability. In order to address this problem, we propose the method of jointly embedding words and labels whereby attention modules learn the weights of words from medical notes according to their relevance to the names of risk prediction labels. This approach boosts interpretability by employing an attention mechanism and including the names of prediction tasks in the model. However, its application is only limited to the handling of textual inputs such as medical notes. In this paper, we propose a label dependent attention model LDAM to 1) improve the interpretability by exploiting Clinical-BERT (a biomedical language model pre-trained on a large clinical corpus) to encode biomedically meaningful features and labels jointly; 2) extend the idea of joint embedding to the processing of time-series data, and develop a multi-modal learning framework for integrating heterogeneous information from medical notes and time-series health status indicators. To demonstrate our method, we apply LDAM to the MIMIC-III dataset to predict different disease risks. We evaluate our method both quantitatively and qualitatively. Specifically, the predictive power of LDAM will be shown, and case studies will be carried out to illustrate its interpretability.

cs.AI↗

arXiv2021

Bayesian data assimilation for estimating epidemic evolution: a COVID-19 study

The evolution of epidemiological parameters, such as instantaneous reproduction number Rt, is important for understanding the transmission dynamics of infectious diseases. Current estimates of time-varying epidemiological parameters often face problems such as lagging observations, averaging inference, and improper quantification of uncertainties. To address these problems, we propose a Bayesian data assimilation framework for time-varying parameter estimation. Specifically, this framework is applied to Rt estimation, resulting in the state-of-the-art DARt system. With DARt, time misalignment caused by lagging observations is tackled by incorporating observation delays into the joint inference of infections and Rt; the drawback of averaging is overcome by instantaneously updating upon new observations and developing a model selection mechanism that captures abrupt changes; the uncertainty is quantified and reduced by employing Bayesian smoothing. We validate the performance of DARt and demonstrate its power in revealing the transmission dynamics of COVID-19. The proposed approach provides a promising solution for accurate and timely estimating transmission dynamics from reported data.

stat.AP↗

arXiv2021

Self-Supervised Detection of Contextual Synonyms in a Multi-Class Setting: Phenotype Annotation Use Case

Contextualised word embeddings is a powerful tool to detect contextual synonyms. However, most of the current state-of-the-art (SOTA) deep learning concept extraction methods remain supervised and underexploit the potential of the context. In this paper, we propose a self-supervised pre-training approach which is able to detect contextual synonyms of concepts being training on the data created by shallow matching. We apply our methodology in the sparse multi-class setting (over 15,000 concepts) to extract phenotype information from electronic health records. We further investigate data augmentation techniques to address the problem of the class sparsity. Our approach achieves a new SOTA for the unsupervised phenotype concept annotation on clinical text on F1 and Recall outperforming the previous SOTA with a gain of up to 4.5 and 4.0 absolute points, respectively. After fine-tuning with as little as 20\% of the labelled data, we also outperform BioBERT and ClinicalBERT. The extrinsic evaluation on three ICU benchmarks also shows the benefit of using the phenotypes annotated by our model as features.

cs.CL↗

arXiv2021

Mitigating Backdoor Attacks in Federated Learning

Malicious clients can attack federated learning systems using malicious data, including backdoor samples, during the training phase. The compromised global model will perform well on the validation dataset designed for the task, but a small subset of data with backdoor patterns may trigger the model to make a wrong prediction. There has been an arms race between attackers who tried to conceal attacks and defenders who tried to detect attacks during the aggregation stage of training on the server-side. In this work, we propose a new and effective method to mitigate backdoor attacks after the training phase. Specifically, we design a federated pruning method to remove redundant neurons in the network and then adjust the model's extreme weight values. Our experiments conducted on distributed Fashion-MNIST show that our method can reduce the average attack success rate from 99.7% to 1.9% with a 5.5% loss of test accuracy on the validation dataset. To minimize the pruning influence on test accuracy, we can fine-tune after pruning, and the attack success rate drops to 6.4%, with only a 1.7% loss of test accuracy. Further experiments under Distributed Backdoor Attacks on CIFAR-10 also show promising results that the average attack success rate drops more than 70% with less than 2% loss of test accuracy on the validation dataset.

cs.CR↗

arXiv2020

An Epidemiological Modelling Approach for Covid19 via Data Assimilation

The global pandemic of the 2019-nCov requires the evaluation of policy interventions to mitigate future social and economic costs of quarantine measures worldwide. We propose an epidemiological model for forecasting and policy evaluation which incorporates new data in real-time through variational data assimilation. We analyze and discuss infection rates in China, the US and Italy. In particular, we develop a custom compartmental SIR model fit to variables related to the epidemic in Chinese cities, named SITR model. We compare and discuss model results which conducts updates as new observations become available. A hybrid data assimilation approach is applied to make results robust to initial conditions. We use the model to do inference on infection numbers as well as parameters such as the disease transmissibility rate or the rate of recovery. The parameterisation of the model is parsimonious and extendable, allowing for the incorporation of additional data and parameters of interest. This allows for scalability and the extension of the model to other locations or the adaption of novel data sources.

stat.AP↗

arXiv2020

A Bayesian Updating Scheme for Pandemics: Estimating the Infection Dynamics of COVID-19

Epidemic models play a key role in understanding and responding to the emerging COVID-19 pandemic. Widely used compartmental models are static and are of limited use to evaluate intervention strategies with the emerging pandemic. Applying the technology of data assimilation, we propose a Bayesian updating approach for estimating epidemiological parameters using observable information for the purpose of assessing the impacts of different intervention strategies. We adopt a concise renewal model and propose new parameters by disentangling the reduction of instantaneous reproduction number Rt into mitigation and suppression factors for quantifying intervention impacts at a finer granularity. Then we developed a data assimilation framework for estimating these parameters including constructing an observation function and developing a Bayesian updating scheme. A statistical analysis framework is then built to quantify the impact of intervention strategies by monitoring the evolution of these estimated parameters. By Investigating the impacts of intervention measures of European countries, the United States and Wuhan with the framework, we reveal the effects of interventions in these countries and the resurgence risk in the USA.

q-bio.PE↗

arXiv2020

Spin-orbit quantum impurity and quantization in a topological magnet

Quantum states induced by single-atomic impurities are at the frontier of physics and material science. While such states have been reported in high-temperature superconductors and dilute magnetic semiconductors, they are unexplored in topological magnets which can feature spin-orbit tunability. Here we use spin-polarized scanning tunneling microscopy/spectroscopy (STM/S) to study the engineered quantum impurity in a topological magnet Co3Sn2S2. We find that each substituted In impurity introduces a striking localized bound state. Our systematic magnetization-polarized probe reveals that this bound state is spin-down polarized, in lock with a negative orbital magnetization. Moreover, the magnetic bound states of neighboring impurities interact to form quantized orbitals, exhibiting an intriguing spin-orbit splitting, analogous to the splitting of the topological fermion line. Our work collectively demonstrates the strong spin-orbit effect of the single-atomic impurity at the quantum level, suggesting that a nonmagnetic impurity can introduce spin-orbit coupled magnetic resonance in topological magnets.

cond-mat.mtrl-sci↗

arXiv2020

Discovery of a quantum limit Chern magnet TbMn6Sn6

The quantum level interplay between geometry, topology, and correlation is at the forefront of fundamental physics. Owing to the unusual lattice geometry and breaking of time-reversal symmetry, kagome magnets are predicted to support intrinsic Chern quantum phases. However, quantum materials hosting ideal spin-orbit coupled kagome lattices with strong out-of-plane magnetization have been lacking. Here we use scanning tunneling microscopy to discover a new topological kagome magnet TbMn6Sn6, which is close to satisfying the above criteria. We visualize its effectively defect-free purely Mn-based ferromagnetic kagome lattice with atomic resolution. Remarkably, its electronic state exhibits distinct Landau quantization upon the application of a magnetic field, and the quantized Landau fan structure features spin-polarized Dirac dispersion with a large Chern gap. We further demonstrate the bulk-boundary correspondence between the Chern gap and topological edge state, as well as the Berry curvature field correspondence of Chern gapped Dirac fermions. Our results point to the realization of a quantum-limit Chern phase in TbMn6Sn6, opening up an avenue for discovering topological quantum phenomena in the RMn6Sn6 (R = rare earth element) family with a variety of magnetic structures. Our visualization of the magnetic bulk-boundary-Berry correspondence covering real and momentum space demonstrates a proof-of-principle method revealing topological magnets.

cond-mat.mtrl-sci↗

arXiv2020

Spin-orbit quantum impurity in a topological kagome magnet

Quantum states induced by single-atomic-impurities are the current frontier of material and information science. Recently the spin-orbit coupled correlated kagome magnets are emerging as a new class of topological quantum materials, although the effect of single-atomic impurities remains unexplored. Here we use state-of-the-art scanning tunneling microscopy/spectroscopy (STM/S) to study the atomic indium impurity in a topological kagome magnet Co3Sn2S2, which is designed to support the spin-orbit quantum state. We find each impurity features a strongly localized bound state. Our systematic magnetization-polarized tunneling probe reveals its spin-down polarized nature with an unusual moment of -5uB, indicative of additional orbital magnetization. As the separation between two impurities progressively shrinks, their respective bound states interact and form quantized molecular orbital states. The molecular orbital of three neighboring impurities further exhibits an intriguing splitting owing to the combination of geometry, magnetism, and spin-orbit coupling, analogous to the splitting of the topological Weyl fermion line12,19. Our work demonstrates the quantum-level interplay between magnetism and spin-orbit coupling at an individual atomic impurity, which provides insights into the emergent impurity behavior in a topological kagome magnet and the potential of spin-orbit quantum impurities for information science.

cond-mat.mtrl-sci↗

arXiv2019

Unsupervised Annotation of Phenotypic Abnormalities via Semantic Latent Representations on Electronic Health Records

The extraction of phenotype information which is naturally contained in electronic health records (EHRs) has been found to be useful in various clinical informatics applications such as disease diagnosis. However, due to imprecise descriptions, lack of gold standards and the demand for efficiency, annotating phenotypic abnormalities on millions of EHR narratives is still challenging. In this work, we propose a novel unsupervised deep learning framework to annotate the phenotypic abnormalities from EHRs via semantic latent representations. The proposed framework takes the advantage of Human Phenotype Ontology (HPO), which is a knowledge base of phenotypic abnormalities, to standardize the annotation results. Experiments have been conducted on 52,722 EHRs from MIMIC-III dataset. Quantitative and qualitative analysis have shown the proposed framework achieves state-of-the-art annotation performance and computational efficiency compared with other methods.

cs.CL↗

arXiv2019

Quantum phase transition of correlated iron-based superconductivity in LiFe$_{1-x}$Co$_x$As

The interplay between unconventional Cooper pairing and quantum states associated with atomic scale defects is a frontier of research with many open questions. So far, only a few of the high-temperature superconductors allow this intricate physics to be studied in a widely tunable way. We use scanning tunneling microscopy (STM) to image the electronic impact of Co atoms on the ground state of the LiFe$_{1-x}$Co$_x$As system. We observe that impurities progressively suppress the global superconducting gap and introduce low energy states near the gap edge, with the superconductivity remaining in the strong-coupling limit. Unexpectedly, the fully opened gap evolves into a nodal state before the Cooper pair coherence is fully destroyed. Our systematic theoretical analysis shows that these new observations can be quantitatively understood by the nonmagnetic Born-limit scattering effect in a s$\pm$-wave superconductor, unveiling the driving force of the superconductor to metal quantum phase transition.

cond-mat.supr-con↗

arXiv2019

Integrated Multi-omics Analysis Using Variational Autoencoders: Application to Pan-cancer Classification

Different aspects of a clinical sample can be revealed by multiple types of omics data. Integrated analysis of multi-omics data provides a comprehensive view of patients, which has the potential to facilitate more accurate clinical decision making. However, omics data are normally high dimensional with large number of molecular features and relatively small number of available samples with clinical labels. The "dimensionality curse" makes it challenging to train a machine learning model using high dimensional omics data like DNA methylation and gene expression profiles. Here we propose an end-to-end deep learning model called OmiVAE to extract low dimensional features and classify samples from multi-omics data. OmiVAE combines the basic structure of variational autoencoders with a classification network to achieve task-oriented feature extraction and multi-class classification. The training procedure of OmiVAE is comprised of an unsupervised phase without the classifier and a supervised phase with the classifier. During the unsupervised phase, a hierarchical cluster structure of samples can be automatically formed without the need for labels. And in the supervised phase, OmiVAE achieved an average classification accuracy of 97.49% after 10-fold cross-validation among 33 tumour types and normal samples, which shows better performance than other existing methods. The OmiVAE model learned from multi-omics data outperformed that using only one type of omics data, which indicates that the complementary information from different omics datatypes provides useful insights for biomedical tasks like cancer classification.

cs.LG↗

arXiv2009

Origin of the X-rays and Possible GeV-TeV Emission from the Western Hot Spot Of Pictor A

Pictor A is a nearby Fanaroff-Riley class II (FR II) radio galaxy with a bright hot spot, the western hot spot. Observation of high polarization in the optical emission of the hot spot indicates that the optical emission could be synchrotron radiation of relativistic electrons in the hot spot. These electrons may be able to produce high energy gamma-ray photons through inverse Compton (IC) scattering. We use single-zone and multi-zone synchrotron + synchrotron-self-Compton (SSC) models to fit the observed spectral energy distribution (SED) from the radio to the X-ray band of the hot spot. Our esults show that in the case of a much weaker magnetic field strength than the equipartition magnetic field, both the single-zone and multi-zone models can fit the SED, but the multi-zone model significantly improves the fit. The two models predict the hot spot as a GeV-TeV source, which might be marginally detectable with Fermi/LAT and HESS. The inverse Compton scattering of cosmic microwave background (IC/CMB) is also considered, but its contribution to GeV-TeV emission is negligible. Note that under the equipartition condition, the SED can also be fit with the multi-zone model, but the predicted flux at 10^22 Hz is too weak to be detectable. The detection of TeV gamma-rays from this FR II radio galaxy, if confirmed, would establish a new subclass of extragalactic source in this energy regime since most of the AGNs detected to date at TeV energies are high-energy-peaked BL Lac objects.

astro-ph.HE↗