arXiv subjects

Ying Ding

Publications and source records attributed to Ying Ding.

At least 109 records · Page 6Linked to original sources

arXiv2021

SCALP -- Supervised Contrastive Learning for Cardiopulmonary Disease Classification and Localization in Chest X-rays using Patient Metadata

Computer-aided diagnosis plays a salient role in more accessible and accurate cardiopulmonary diseases classification and localization on chest radiography. Millions of people get affected and die due to these diseases without an accurate and timely diagnosis. Recently proposed contrastive learning heavily relies on data augmentation, especially positive data augmentation. However, generating clinically-accurate data augmentations for medical images is extremely difficult because the common data augmentation methods in computer vision, such as sharp, blur, and crop operations, can severely alter the clinical settings of medical images. In this paper, we proposed a novel and simple data augmentation method based on patient metadata and supervised knowledge to create clinically accurate positive and negative augmentations for chest X-rays. We introduce an end-to-end framework, SCALP, which extends the self-supervised contrastive approach to a supervised setting. Specifically, SCALP pulls together chest X-rays from the same patient (positive keys) and pushes apart chest X-rays from different patients (negative keys). In addition, it uses ResNet-50 along with the triplet-attention mechanism to identify cardiopulmonary diseases, and Grad-CAM++ to highlight the abnormal regions. Our extensive experiments demonstrate that SCALP outperforms existing baselines with significant margins in both classification and localization tasks. Specifically, the average classification AUCs improve from 82.8% (SOTA using DenseNet-121) to 83.9% (SCALP using ResNet-50), while the localization results improve on average by 3.7% over different IoU thresholds.

eess.IV↗

arXiv2021

Spending Your Winning Lottery Better After Drawing It

Lottery Ticket Hypothesis (LTH) suggests that a dense neural network contains a sparse sub-network that can match the performance of the original dense network when trained in isolation from scratch. Most works retrain the sparse sub-network with the same training protocols as its dense network, such as initialization, architecture blocks, and training recipes. However, till now it is unclear that whether these training protocols are optimal for sparse networks. In this paper, we demonstrate that it is unnecessary for spare retraining to strictly inherit those properties from the dense network. Instead, by plugging in purposeful "tweaks" of the sparse subnetwork architecture or its training recipe, its retraining can be significantly improved than the default, especially at high sparsity levels. Combining all our proposed "tweaks" can yield the new state-of-the-art performance of LTH, and these modifications can be easily adapted to other sparse training algorithms in general. Specifically, we have achieved a significant and consistent performance gain of1.05% - 4.93% for ResNet18 on CIFAR-100 over vanilla-LTH. Moreover, our methods are shown to generalize across datasets (CIFAR10, CIFAR100, TinyImageNet) and architectures (Vgg16, ResNet-18/ResNet-34, MobileNet). All codes will be publicly available.

cs.LG↗

arXiv2021

Using Radiomics as Prior Knowledge for Thorax Disease Classification and Localization in Chest X-rays

Chest X-ray becomes one of the most common medical diagnoses due to its noninvasiveness. The number of chest X-ray images has skyrocketed, but reading chest X-rays still have been manually performed by radiologists, which creates huge burnouts and delays. Traditionally, radiomics, as a subfield of radiology that can extract a large number of quantitative features from medical images, demonstrates its potential to facilitate medical imaging diagnosis before the deep learning era. In this paper, we develop an end-to-end framework, ChexRadiNet, that can utilize the radiomics features to improve the abnormality classification performance. Specifically, ChexRadiNet first applies a light-weight but efficient triplet-attention mechanism to classify the chest X-rays and highlight the abnormal regions. Then it uses the generated class activation map to extract radiomic features, which further guides our model to learn more robust image features. After a number of iterations and with the help of radiomic features, our framework can converge to more accurate image regions. We evaluate the ChexRadiNet framework using three public datasets: NIH ChestX-ray, CheXpert, and MIMIC-CXR. We find that ChexRadiNet outperforms the state-of-the-art on both disease detection (0.843 in AUC) and localization (0.679 in T(IoU) = 0.1). We will make the code publicly available at https://github.com/bionlplab/lung_disease_detection_amia2021, with the hope that this method can facilitate the development of automatic systems with a higher-level understanding of the radiological world.

cs.CV↗

arXiv2021

A dataset of mentorship in science with semantic and demographic estimations

Mentorship in science is crucial for topic choice, career decisions, and the success of mentees and mentors. Typically, researchers who study mentorship use article co-authorship and doctoral dissertation datasets. However, available datasets of this type focus on narrow selections of fields and miss out on early career and non-publication-related interactions. Here, we describe MENTORSHIP, a crowdsourced dataset of 743176 mentorship relationships among 738989 scientists across 112 fields that avoids these shortcomings. We enrich the scientists' profiles with publication data from the Microsoft Academic Graph and "semantic" representations of research using deep learning content analysis. Because gender and race have become critical dimensions when analyzing mentorship and disparities in science, we also provide estimations of these factors. We perform extensive validations of the profile--publication matching, semantic content, and demographic inferences. We anticipate this dataset will spur the study of mentorship in science and deepen our understanding of its role in scientists' career outcomes.

cs.DL↗

arXiv2021

Bootstrapping Your Own Positive Sample: Contrastive Learning With Electronic Health Record Data

Electronic Health Record (EHR) data has been of tremendous utility in Artificial Intelligence (AI) for healthcare such as predicting future clinical events. These tasks, however, often come with many challenges when using classical machine learning models due to a myriad of factors including class imbalance and data heterogeneity (i.e., the complex intra-class variances). To address some of these research gaps, this paper leverages the exciting contrastive learning framework and proposes a novel contrastive regularized clinical classification model. The contrastive loss is found to substantially augment EHR-based prediction: it effectively characterizes the similar/dissimilar patterns (by its "push-and-pull" form), meanwhile mitigating the highly skewed class distribution by learning more balanced feature spaces (as also echoed by recent findings). In particular, when naively exporting the contrastive learning to the EHR data, one hurdle is in generating positive samples, since EHR data is not as amendable to data augmentation as image data. To this end, we have introduced two unique positive sampling strategies specifically tailored for EHR data: a feature-based positive sampling that exploits the feature space neighborhood structure to reinforce the feature learning; and an attribute-based positive sampling that incorporates pre-generated patient similarity metrics to define the sample proximity. Both sampling approaches are designed with an awareness of unique high intra-class variance in EHR data. Our overall framework yields highly competitive experimental results in predicting the mortality risk on real-world COVID-19 EHR data with a total of 5,712 patients admitted to a large, urban health system. Specifically, our method reaches a high AUROC prediction score of 0.959, which outperforms other baselines and alternatives: cross-entropy(0.873) and focal loss(0.931).

cs.LG↗

arXiv2021

Innovation adoption: Broadcasting vs. Virality

Diffusion channels are critical to determining the adoption scale which leads to the ultimate impact of an innovation. The aim of this study is to develop an integrative understanding of the impact of two diffusion channels (i.e., broadcasting vs virality) on innovation adoption. Using citations of a series of classic algorithms and the time series of co-authorship as the footprints of their diffusion trajectories, we propose a novel method to analyze the intertwining relationships between broadcasting and virality in the innovation diffusion process. Our findings show that broadcasting and virality have similar diffusion power, but play different roles across diffusion stages. Broadcasting is more powerful in the early stages but may be gradually caught up or even surpassed by virality in the later period. Meanwhile, diffusion speed in virality is significantly faster than broadcasting and members from virality channels tend to adopt the same innovation repetitively.

cs.SI↗

arXiv2021

Evolution of Charge and Pair Density Modulations in Overdoped Bi2Sr2CuO6+delta

One of the central issues concerning the mechanism of high temperature superconductivity in cuprates is the nature of the ubiquitous charge order and its implications to superconductivity. Here we use scanning tunneling microscopy to investigate the evolution of charge order from the optimally doped to strongly overdoped Bi2Sr2CuO6+δ cuprates. We find that with increasing hole concentration, the long-range checkerboard order gradually evolves into short-range glassy patterns consisting of diluted charge puddles. Each charge puddle has a unidirectional nematic internal structure, and exhibits clear pair density modulations as revealed by the spatial variations of superconducting coherence peak and gap depth. Both the charge puddles and the nematicity vanish completely in the strongly overdoped non-superconducting regime, when another type of short-range order with root2 * root2 periodicity emerges. These results shed important new lights on the intricate interplay between the intertwined orders and the superconducting phase of cuprates.

cond-mat.supr-con↗

arXiv2021

Contrastive Learning Improves Critical Event Prediction in COVID-19 Patients

Machine Learning (ML) models typically require large-scale, balanced training data to be robust, generalizable, and effective in the context of healthcare. This has been a major issue for developing ML models for the coronavirus-disease 2019 (COVID-19) pandemic where data is highly imbalanced, particularly within electronic health records (EHR) research. Conventional approaches in ML use cross-entropy loss (CEL) that often suffers from poor margin classification. For the first time, we show that contrastive loss (CL) improves the performance of CEL especially for imbalanced EHR data and the related COVID-19 analyses. This study has been approved by the Institutional Review Board at the Icahn School of Medicine at Mount Sinai. We use EHR data from five hospitals within the Mount Sinai Health System (MSHS) to predict mortality, intubation, and intensive care unit (ICU) transfer in hospitalized COVID-19 patients over 24 and 48 hour time windows. We train two sequential architectures (RNN and RETAIN) using two loss functions (CEL and CL). Models are tested on full sample data set which contain all available data and restricted data set to emulate higher class imbalance.CL models consistently outperform CEL models with the restricted data set on these tasks with differences ranging from 0.04 to 0.15 for AUPRC and 0.05 to 0.1 for AUROC. For the restricted sample, only the CL model maintains proper clustering and is able to identify important features, such as pulse oximetry. CL outperforms CEL in instances of severe class imbalance, on three EHR outcomes with respect to three performance metrics: predictive power, clustering, and feature importance. We believe that the developed CL framework can be expanded and used for EHR ML work in general.

cs.LG↗

arXiv2020

Deep Learning with Heterogeneous Graph Embeddings for Mortality Prediction from Electronic Health Records

Computational prediction of in-hospital mortality in the setting of an intensive care unit can help clinical practitioners to guide care and make early decisions for interventions. As clinical data are complex and varied in their structure and components, continued innovation of modeling strategies is required to identify architectures that can best model outcomes. In this work, we train a Heterogeneous Graph Model (HGM) on Electronic Health Record data and use the resulting embedding vector as additional information added to a Convolutional Neural Network (CNN) model for predicting in-hospital mortality. We show that the additional information provided by including time as a vector in the embedding captures the relationships between medical concepts, lab tests, and diagnoses, which enhances predictive performance. We find that adding HGM to a CNN model increases the mortality prediction accuracy up to 4\%. This framework serves as a foundation for future experiments involving different EHR data types on important healthcare prediction tasks.

cs.LG↗

arXiv2020

Understanding Team Collaboration in Artificial Intelligence from the perspective of Geographic Distance

This paper analyzes team collaboration in the field of Artificial Intelligence (AI) from the perspective of geographic distance. We obtained 1,584,175 AI related publications during 1950-2019 from the Microsoft Academic Graph. Three latitude-and-longitude-based indicators were employed to quantify the geographic distance of collaborations in AI over time at domestic and international levels. The results show team collaborations in AI has been more popular in the field over time with around 42,000 (38.4%) multiple-affiliation AI publications in 2019. The changes in geographic distances of team collaborations indicate the increase of breadth and density for both domestic and international collaborations in AI over time. In addition, the United States produced the largest number of single-country and internationally collaborated AI publications, and China has played an important role in international collaborations in AI after 2010.

cs.DL↗

arXiv2020

Biomedical Knowledge Graph Refinement and Completion using Graph Representation Learning and Top-K Similarity Measure

Knowledge Graphs have been one of the fundamental methods for integrating heterogeneous data sources. Integrating heterogeneous data sources is crucial, especially in the biomedical domain, where central data-driven tasks such as drug discovery rely on incorporating information from different biomedical databases. These databases contain various biological entities and relations such as proteins (PDB), genes (Gene Ontology), drugs (DrugBank), diseases (DDB), and protein-protein interactions (BioGRID). The process of semantically integrating heterogeneous biomedical databases is often riddled with imperfections. The quality of data-driven drug discovery relies on the accuracy of the mining methods used and the data's quality as well. Thus, having complete and refined biomedical knowledge graphs is central to achieving more accurate drug discovery outcomes. Here we propose using the latest graph representation learning and embedding models to refine and complete biomedical knowledge graphs. This preliminary work demonstrates learning discrete representations of the integrated biomedical knowledge graph Chem2Bio2RD [3]. We perform a knowledge graph completion and refinement task using a simple top-K cosine similarity measure between the learned embedding vectors to predict missing links between drugs and targets present in the data. We show that this simple procedure can be used alternatively to binary classifiers in link prediction.

cs.LG↗

arXiv2020

Exploring Direct Citations between Citing Publications

This paper defines and explores the direct citations between citing publications (DCCPs) of a publication. We construct an ego-centered citation network for each paper that contains all of its citing papers and itself, as well as the citation relationships among them. By utilizing a large-scale scholarly dataset from the computer science field in the Microsoft Academic Graph (MAG-CS) dataset, we find that DCCPs exist universally in medium and highly cited papers. For those papers that have DCCPs, DCCPs do occur frequently; highly cited papers tend to contain more DCCPs than others. Meanwhile, the number of DCCPs of papers published in different years does not vary dramatically. The current paper also discusses the relationship between DCCPs and some indirect citation relationships (e.g., co-citation and bibliographic coupling).

cs.DL↗

arXiv2020

The Pace of Artificial Intelligence Innovations: Speed, Talent, and Trial-and-Error

Innovations in artificial intelligence (AI) are occurring at speeds faster than ever witnessed before. However, few studies have managed to measure or depict this increasing velocity of innovations in the field of AI. In this paper, we combine data on AI from arXiv and Semantic Scholar to explore the pace of AI innovations from three perspectives: AI publications, AI players, and AI updates (trial and error). A research framework and three novel indicators, Average Time Interval (ATI), Innovation Speed (IS) and Update Speed (US), are proposed to measure the pace of innovations in the field of AI. The results show that: (1) in 2019, more than 3 AI preprints were submitted to arXiv per hour, over 148 times faster than in 1994. Furthermore, there was one deep learning-related preprint submitted to arXiv every 0.87 hours in 2019, over 1,064 times faster than in 1994. (2) For AI players, 5.26 new researchers entered into the field of AI each hour in 2019, more than 175 times faster than in the 1990s. (3) As for AI updates (trial and error), one updated AI preprint was submitted to arXiv every 41 days, with around 33% of AI preprints having been updated at least twice in 2019. In addition, as reported in 2019, it took, on average, only around 0.2 year for AI preprints to receive their first citations, which is 5 times faster than 2000-2007. This swift pace in AI illustrates the increase in popularity of AI innovation. The systematic and fine-grained analysis of the AI field enabled to portrait the pace of AI innovation and demonstrated that the proposed approach can be adopted to understand other fast-growing fields such as cancer research and nano science.

cs.DL↗

arXiv2020

Coronavirus Knowledge Graph: A Case Study

The emergence of the novel COVID-19 pandemic has had a significant impact on global healthcare and the economy over the past few months. The virus's rapid widespread has led to a proliferation in biomedical research addressing the pandemic and its related topics. One of the essential Knowledge Discovery tools that could help the biomedical research community understand and eventually find a cure for COVID-19 are Knowledge Graphs. The CORD-19 dataset is a collection of publicly available full-text research articles that have been recently published on COVID-19 and coronavirus topics. Here, we use several Machine Learning, Deep Learning, and Knowledge Graph construction and mining techniques to formalize and extract insights from the PubMed dataset and the CORD-19 dataset to identify COVID-19 related experts and bio-entities. Besides, we suggest possible techniques to predict related diseases, drug candidates, gene, gene mutations, and related compounds as part of a systematic effort to apply Knowledge Discovery methods to help biomedical researchers tackle the pandemic.

cs.AI↗

arXiv2020

Analysis of misinformation during the COVID-19 outbreak in China: cultural, social and political entanglements

COVID-19 resulted in an infodemic, which could erode public trust, impede virus containment, and outlive the pandemic itself. The evolving and fragmented media landscape is a key driver of the spread of misinformation. Using misinformation identified by the fact-checking platform by Tencent and posts on Weibo, our results showed that the evolution of misinformation follows an issue-attention cycle, pertaining to topics such as city lockdown, cures, and preventions, and school reopening. Sources of authority weigh in on these topics, but their influence is complicated by peoples' pre-existing beliefs and cultural practices. Finally, social media has a complicated relationship with established or legacy media systems. Sometimes they reinforce each other, but in general, social media may have a topic cycle of its own making. Our findings shed light on the distinct characteristics of misinformation during the COVID-19 and offer insights into combating misinformation in China and across the world at large.

cs.SI↗

arXiv2020

Building a PubMed knowledge graph

PubMed is an essential resource for the medical domain, but useful concepts are either difficult to extract or are ambiguated, which has significantly hindered knowledge discovery. To address this issue, we constructed a PubMed knowledge graph (PKG) by extracting bio-entities from 29 million PubMed abstracts, disambiguating author names, integrating funding data through the National Institutes of Health (NIH) ExPORTER, collecting affiliation history and educational background of authors from ORCID, and identifying fine-grained affiliation data from MapAffil. Through the integration of the credible multi-source data, we could create connections among the bio-entities, authors, articles, affiliations, and funding. Data validation revealed that the BioBERT deep learning method of bio-entity extraction significantly outperformed the state-of-the-art models based on the F1 score (by 0.51%), with the author name disambiguation (AND) achieving a F1 score of 98.09%. PKG can trigger broader innovations, not only enabling us to measure scholarly impact, knowledge usage, and knowledge transfer, but also assisting us in profiling authors and organizations based on their connections with bio-entities. The PKG is freely available on Figshare (https://figshare.com/s/6327a55355fc2c99f3a2, simplified version that exclude PubMed raw data) and TACC website (http://er.tacc.utexas.edu/datasets/ped, full version).

cs.DL↗

arXiv2020

Attribute2vec: Deep Network Embedding Through Multi-Filtering GCN

We present a multi-filtering Graph Convolution Neural Network (GCN) framework for network embedding task. It uses multiple local GCN filters to do feature extraction in every propagation layer. We show this approach could capture different important aspects of node features against the existing attribute embedding based method. We also show that with multi-filtering GCN approach, we can achieve significant improvement against baseline methods when training data is limited. We also perform many empirical experiments and demonstrate the benefit of using multiple filters against single filter as well as most current existing network embedding methods for both the link prediction and node classification tasks.

cs.LG↗

arXiv2020

A Simultaneous Inference Procedure to Identify Subgroups from RCTs with Survival Outcomes: Application to Analysis of AMD Progression Studies

With the uptake of targeted therapies, instead of the "one-fits-all" approach, modern randomized clinical trials (RCTs) often aim to develop treatments that target a subgroup of patients. Motivated by analyzing the Age-Related Eye Disease Study (AREDS) data, a large RCT to study the efficacy of nutritional supplements in delaying the progression of an eye disease, age-related macular degeneration (AMD), we develop a simultaneous inference procedure to identify and infer subgroups with differential treatment efficacy in RCTs with survival outcome. Specifically, we formulate the multiple testing problem through contrasts and construct their simultaneous confidence intervals, which control both within- and across- marker multiplicity appropriately. Realistic simulations are conducted using real genotype data to evaluate the method performance under various scenarios. The method is then applied to AREDS to assess the efficacy of antioxidants and zinc combination in delaying AMD progression. Multiple gene regions including ESRRB-VASH1 on chromosome 14 have been identified with subgroups showing differential efficacy. We further validate our findings in an independent subsequent RCT, AREDS2, by discovering consistent differential treatment responses in the targeted and non-targeted subgroups been identified from AREDS. This simultaneous inference approach provides a step forward to confidently identify and infer subgroups in modern drug development.

stat.AP↗