One major impediment to the wider use of deep learning for clinical decision making is the difficulty of assigning a level of confidence to model predictions. Currently, deep Bayesian neural networks and sparse Gaussian processes are the main two scalable uncertainty estimation methods. However, deep Bayesian neural network suffers from lack of expressiveness, and more expressive models such as deep kernel learning, which is an extension of sparse Gaussian process, captures only the uncertainty from the higher level latent space. Therefore, the deep learning model under it lacks interpretability and ignores uncertainty from the raw data. In this paper, we merge features of the deep Bayesian learning framework with deep kernel learning to leverage the strengths of both methods for more comprehensive uncertainty estimation. Through a series of experiments on predicting the first incidence of heart failure, diabetes and depression applied to large-scale electronic medical records, we demonstrate that our method is better at capturing uncertainty than both Gaussian processes and deep Bayesian neural networks in terms of indicating data insufficiency and distinguishing true positive and false positive predictions, with a comparable generalisation performance. Furthermore, by assessing the accuracy and area under the receiver operating characteristic curve over the predictive probability, we show that our method is less susceptible to making overconfident predictions, especially for the minority class in imbalanced datasets. Finally, we demonstrate how uncertainty information derived by the model can inform risk factor analysis towards model interpretability.
Generative Adversarial Networks (GANs) represent a promising class of generative networks that combine neural networks with game theory. From generating realistic images and videos to assisting musical creation, GANs are transforming many fields of arts and sciences. However, their application to healthcare has not been fully realized, more specifically in generating electronic health records (EHR) data. In this paper, we propose a framework for exploring the value of GANs in the context of continuous laboratory time series data. We devise an unsupervised evaluation method that measures the pr…
University of Southern California（南加州大学）；AI Foundations, IBM Thomas J. Watson Research Center（IBM托马斯·J·沃森研究中心AI基础部门）；Binghamton University, SUNY（纽约州立大学宾汉姆顿大学）；IBM Thomas J. Watson Research Center（IBM托马斯·J·沃森研究中心）
The rapid growth of Electronic Health Records (EHRs), as well as the accompanied opportunities in Data-Driven Healthcare (DDH), has been attracting widespread interests and attentions. Recent progress in the design and applications of deep learning methods has shown promising results and is forcing massive changes in healthcare academia and industry, but most of these methods rely on massive labeled data. In this work, we propose a general deep learning framework which is able to boost risk prediction performance with limited EHR data. Our model takes a modified generative adversarial network …
Virginia Polytechnic Institute and State University (Virginia Tech)（弗吉尼亚理工学院暨州立大学（弗吉尼亚理工））
Deep learning models have demonstrated high-quality performance in areas such as image classification and speech processing. However, creating a deep learning model using electronic health record (EHR) data, requires addressing particular privacy challenges that are unique to researchers in this domain. This matter focuses attention on generating realistic synthetic data while ensuring privacy. In this paper, we propose a novel framework called correlation-capturing Generative Adversarial Network (CorGAN), to generate synthetic healthcare records. In CorGAN we utilize Convolutional Neural Netw…
University of North Carolina at Charlotte（北卡罗来纳大学夏洛特分校）
Deep learning models have exhibited superior performance in predictive tasks with the explosively increasing Electronic Health Records (EHR). However, due to the lack of transparency, behaviors of deep learning models are difficult to interpret. Without trustworthiness, deep learning models will not be able to assist in the real-world decision-making process of healthcare issues. We propose a deep learning model based on Bayesian Neural Networks (BNN) to predict uncertainty induced by data noise. The uncertainty is introduced to provide model predictions with an extra level of confidence. Our …
University of Cambridge（剑桥大学）；The Alan Turing Institute（艾伦·图灵研究所）；University of California Los Angeles（加州大学洛杉矶分校）
Analyzing electronic health records (EHR) poses significant challenges because often few samples are available describing a patient's health and, when available, their information content is highly diverse. The problem we consider is how to integrate sparsely sampled longitudinal data, missing measurements informative of the underlying health status and fixed demographic information to produce estimated survival distributions updated through a patient's follow up. We propose a nonparametric probabilistic model that generates survival trajectories from an ensemble of Bayesian trees that learns …
Centers for Disease Control and Prevention（美国疾病控制与预防中心）
A variety of methods existing for generating synthetic electronic health records (EHRs), but they are not capable of generating unstructured text, like emergency department (ED) chief complaints, history of present illness or progress notes. Here, we use the encoder-decoder model, a deep learning algorithm that features in many contemporary machine translation systems, to generate synthetic chief complaints from discrete variables in EHRs, like age group, gender, and discharge diagnosis. After being trained end-to-end on authentic records, the model can generate realistic chief complaint text …
Inpatient care is a large share of total health care spending, making analysis of inpatient utilization patterns an important part of understanding what drives health care spending growth. Common features of inpatient utilization measures include zero inflation, over-dispersion, and skewness, all of which complicate statistical modeling. Mixture modeling is a popular approach that can accommodate these features of health care utilization data. In this work, we add a nonparametric clustering component to such models. Our fully Bayesian model framework allows for an unknown number of mixing comp…
Kyowa Hakko Bio Co., Ltd.（协和发酵生物株式会社）；Kyoto University（京都大学）；Hirosaki University（弘前大学）
Clinical decision making regarding treatments based on personal characteristics leads to effective health improvements. Machine learning (ML) has been the primary concern of diagnosis support according to comprehensive patient information. However, the remaining prominent issue is the development of objective treatment processes in clinical situations. This study proposes a novel framework to plan treatment processes in a data-driven manner. A key point of the framework is the evaluation of the "actionability" for personal health improvements by using a surrogate Bayesian model in addition to …
We propose multivariate nonstationary Gaussian processes for jointly modeling multiple clinical variables, where the key parameters, length-scales, standard deviations and the correlations between the observed output, are all time dependent. We perform posterior inference via Hamiltonian Monte Carlo (HMC). We also provide methods for obtaining computationally efficient gradient-based maximum a posteriori (MAP) estimates. We validate our model on synthetic data as well as on electronic health records (EHR) data from Kaiser Permanente (KP). We show that the proposed model provides better predict…
Due to patient privacy protection concerns, machine learning research in healthcare has been undeniably slower and limited than in other application domains. High-quality, realistic, synthetic electronic health records (EHRs) can be leveraged to accelerate methodological developments for research purposes while mitigating privacy concerns associated with data sharing. The current state-of-the-art model for synthetic EHR generation is generative adversarial networks, which are notoriously difficult to train and can suffer from mode collapse. Denoising Diffusion Probabilistic Models, a class of …
Vanderbilt University Medical Center（范德比尔特大学医学中心）；Sage Bionetworks；Vanderbilt University（范德比尔特大学）；University of Washington（华盛顿大学）；Tempus Labs
Synthetic health data have the potential to mitigate privacy concerns when sharing data to support biomedical research and the development of innovative healthcare applications. Modern approaches for data generation based on machine learning, generative adversarial networks (GAN) methods in particular, continue to evolve and demonstrate remarkable potential. Yet there is a lack of a systematic assessment framework to benchmark methods as they emerge and determine which methods are most appropriate for which use cases. In this work, we introduce a generalizable benchmarking framework to apprais…
Electronic health records (EHRs) are increasingly used for clinical and comparative effectiveness research, but suffer from missing data. Motivated by health services research on diabetes care, we seek to increase the quality of EHRs by focusing on missing values of longitudinal glycosylated hemoglobin (A1c), a key risk factor for diabetes complications and adverse events. Under the framework of multiple imputation (MI), we propose an individualized Bayesian latent profiling approach to capture A1c measurement trajectories subject to missingness. The proposed method is applied to EHRs of adult…
Probabilistic record linkage (PRL) is the process of determining which records in two databases correspond to the same underlying entity in the absence of a unique identifier. Bayesian solutions to this problem provide a powerful mechanism for propagating uncertainty due to uncertain links between records (via the posterior distribution). However, computational considerations severely limit the practical applicability of existing Bayesian approaches. We propose a new computational approach, providing both a fast algorithm for deriving point estimates of the linkage structure that properly acco…
The analysis of population-wide datasets can provide insight on the health status of large populations so that public health officials can make data-driven decisions. The analysis of such datasets often requires highly parameterized models with different types of fixed and randoms effects to account for risk factors, spatial and temporal variations, multilevel effects and other sources on uncertainty. To illustrate the potential of Bayesian hierarchical models, a dataset of about 500 000 inhabitants released by the Polish National Health Fund containing information about ischemic stroke incide…
This work proposes a new methodology to fit zero inflated Bernoulli data from a Bayesian approach, able to distinguish between two potential sources of zeros (structurals and non-structurals). Its usage is illustrated by means of a real example from the field of occupational health as the phenomenon of sickness presenteeism, in which it is reasonable to think that some individuals will never be at risk of suffering it because they have not been sick in the period of study (structural zeros). Without separating structural and non-structural zeros one would one would be studying jointly the gene…
Researchers are often interested in linking individuals between two datasets that lack a common unique identifier. Matching procedures often struggle to match records with common names, birthplaces or other field values. Computational feasibility is also a challenge, particularly when linking large datasets. We develop a Bayesian method for automated probabilistic record linkage and show it recovers more than 50% more true matches, holding accuracy constant, than comparable methods in a matching of military recruitment data to the 1900 US Census for which expert-labelled matches are available.…
Small area population estimates are useful for decision making in the private and public sectors. However, in small areas (i.e., those that are difficult to reach and with small population sizes) computing demographic quantities is complicated. Bayesian methods are an alternative for demographic population estimates which uses data from multiple administrative records. In this paper we explore a Bayesian approach which is simple and flexible and represents an alternative procedure for base population estimates particularly powerful for intercensal periods. The applicability of the methodologic…
Racialized economic segregation, a key metric that simultaneously accounts for spatial, social and income polarization, has been linked to adverse health outcomes, including morbidity and mortality; however, statistical methods for measuring the association between racialized economic segregation and health outcomes are not well-developed and are usually studied at the individual level. In this paper we propose a two-stage Bayesian statistical framework that provides a broad, flexible approach to studying the spatially varying association between premature mortality and racialized economic seg…
Record linkage (de-duplication or entity resolution) is the process of merging noisy databases to remove duplicate entities. While record linkage removes duplicate entities from such databases, the downstream task is any inferential, predictive, or post-linkage task on the linked data. One goal of the downstream task is obtaining a larger reference data set, allowing one to perform more accurate statistical analyses. In addition, there is inherent record linkage uncertainty passed to the downstream task. Motivated by the above, we propose a generalized Bayesian record linkage method and consid…
Recently the patient centered medical home (PCMH) model has become a popular approach to deliver better care to patients. Current research shows that the most important key for succession of this method is to make balance between healthcare supply and demand. Without such balance in clinical supply and demand, issues such as excessive under and over utilization of physicians, long waiting time for receiving the appropriate treatment, and non continuity of care will eliminate many advantages of the medical home strategy. To reach this end we need to have information about both supply and demand…
The relationship between short-term exposure to air pollution and mortality or morbidity has been the subject of much recent research, in which the standard method of analysis uses Poisson linear or additive models. In this paper we use a Bayesian dynamic generalised linear model (DGLM) to estimate this relationship, which allows the standard linear or additive model to be extended in two ways: (i) the long-term trend and temporal correlation present in the health data can be modelled by an autoregressive process rather than a smooth function of calendar time; (ii) the effects of air pollution…
Improving health worldwide will require rigorous quantification of population-level trends in health status. However, global-level surveys are not available, forcing researchers to rely on fragmentary country-specific data of varying quality. We present a Bayesian model that systematically combines disparate data to make country-, region- and global-level estimates of time trends in important health indicators. The model allows for time and age nonlinearity, and it borrows strength in time, age, covariates, and within and across regional country clusters to make estimates where data are sparse…
We propose a categorical matrix factorization method to infer latent diseases from electronic health records (EHR) data in an unsupervised manner. A latent disease is defined as an unknown biological aberration that causes a set of common symptoms for a group of patients. The proposed approach is based on a novel double feature allocation model which simultaneously allocates features to the rows and the columns of a categorical matrix. Using a Bayesian approach, available prior information on known diseases greatly improves identifiability and interpretability of latent diseases. This includes…
Probabilistic record linkage (PRL) is the process of determining which records in two databases correspond to the same underlying entity in the absence of a unique identifier. Bayesian solutions to this problem provide a powerful mechanism for propagating uncertainty due to uncertain links between records (via the posterior distribution). However, computational considerations severely limit the practical applicability of existing Bayesian approaches. We propose a new computational approach, providing both a fast algorithm for deriving point estimates of the linkage structure that properly acco…
We propose a novel unsupervised approach for linking records across arbitrarily many files, while simultaneously detecting duplicate records within files. Our key innovation is to represent the pattern of links between records as a {\em bipartite} graph, in which records are directly linked to latent true individuals, and only indirectly linked to other records. This flexible new representation of the linkage structure naturally allows us to estimate the attributes of the unique observable people in the population, calculate $k$-way posterior probabilities of matches across records, and propag…
Background. Health economic evaluations of interventions against infectious diseases are commonly based on the predictions of ordinary differential equation (ODE) systems or Markov models (MMs). Standard MMs are static, whereas ODE systems are usually dynamic and account for herd immunity which is crucial to prevent overestimation of infection prevalence. Complex ODE systems including probabilistic model parameters are computationally intensive. Thus, mainly ODE-based models including deterministic parameters are presented in the literature. These do not account for parameter uncertainty. As a…
In epidemiological surveys, data missing not at random (MNAR) due to survey nonresponse may potentially lead to a bias in the risk factor estimates. We propose an approach based on Bayesian data augmentation and survival modelling to reduce the nonresponse bias. The approach requires additional information based on follow-up data. We present a case study of smoking prevalence using FINRISK data collected between 1972 and 2007 with a follow-up to the end of 2012 and compare it to other commonly applied missing at random (MAR) imputation approaches. A simulation experiment is carried out to stud…
The problem we consider considers estimating a multivariate longitudinal panel data model whose outcomes can be a combination of discrete and continuous variables. This problem is challenging because the likelihood is usually analytically intractable. Our article makes both a methodological contribution and also a substantive contribution to the application. The methodological contribution is to introduce into the panel data literature a particle Metropolis within Gibbs method to carry out Bayesian inference, using a Hamiltonian Monte Carlo (Neal 2011} proposal for sampling the vector of unkno…
In electronic health records (EHRs), latent subgroups of patients may exhibit distinctive patterning in their longitudinal health trajectories. For such data, growth mixture models (GMMs) enable classifying patients into different latent classes based on individual trajectories and hypothesized risk factors. However, the application of GMMs is hindered by the special missing data problem in EHRs, which manifests two patient-led missing data processes: the visit process and the response process for an EHR variable conditional on a patient visiting the clinic. If either process is associated wit…
Studies of the relationships between environmental exposures and adverse health outcomes often rely on a two-stage statistical modeling approach, where exposure is modeled/predicted in the first stage and used as input to a separately fit health outcome analysis in the second stage. Uncertainty in these predictions is frequently ignored, or accounted for in an overly simplistic manner, when estimating the associations of interest. Working in the Bayesian setting, we propose a flexible kernel density estimation (KDE) approach for fully utilizing posterior output from the first stage modeling/pr…
Quantifying the value of the information extracted from a structural health monitoring (SHM) system is an important step towards convincing decision makers to implement these systems. We quantify this value by adaptation of the Bayesian decision analysis framework. In contrast to previous works, we model in detail the entire process of data generation to processing, model updating and reliability calculation, and investigate it on a deteriorating bridge system. The framework assumes that dynamic response data are obtained in a sequential fashion from deployed accelerometers, subsequently proce…
Humans are routinely exposed to mixtures of chemical and other environmental factors, making the quantification of health effects associated with environmental mixtures a critical goal for establishing environmental policy sufficiently protective of human health. The quantification of the effects of exposure to an environmental mixture poses several statistical challenges. It is often the case that exposure to multiple pollutants interact with each other to affect an outcome. Further, the exposure-response relationship between an outcome and some exposures, such as some metals, can exhibit com…
Understanding the association between injury severity and patients' potential for recovery is crucial to providing better care for patients with traumatic brain injury (TBI). Estimation of this relationship requires clinical information on injury severity, patient demographics, and healthcare utilization, which are often obtained from separate data sources. Because of privacy and confidentiality regulations, these data sources do not include unique identifiers to link records across data sources. Record linkage is a process to identify records that represent the same entity across data sources…
Estimation of the long-term health effects of air pollution is a challenging task, especially when modelling small-area disease incidence data in an ecological study design. The challenge comes from the unobserved underlying spatial correlation structure in these data, which is accounted for using random effects modelled by a globally smooth conditional autoregressive model. These smooth random effects confound the effects of air pollution, which are also globally smooth. To avoid this collinearity a Bayesian localised conditional autoregressive model is developed for the random effects. This …
Trial-based economic evaluations are typically performed on cross-sectional variables, derived from the responses for only the completers in the study, using methods that ignore the complexities of utility and cost data (e.g. skewness and spikes). We present an alternative and more efficient Bayesian parametric approach to handle missing longitudinal outcomes in economic evaluations, while accounting for the complexities of the data. We specify a flexible parametric model for the observed data and partially identify the distribution of the missing data with partial identifying restrictions and…
Electronic health records (EHR) are rich heterogeneous collection of patient health information, whose broad adoption provides great opportunities for systematic health data mining. However, heterogeneous EHR data types and biased ascertainment impose computational challenges. Here, we present mixEHR, an unsupervised generative model integrating collaborative filtering and latent topic models, which jointly models the discrete distributions of data observation bias and actual data using latent disease-topic distributions. We apply mixEHR on 12.8 million phenotypic observations from the MIMIC d…
The past decade has seen an explosion in the amount of digital information stored in electronic health records (EHR). While primarily designed for archiving patient clinical information and administrative healthcare tasks, many researchers have found secondary use of these records for various clinical informatics tasks. Over the same period, the machine learning community has seen widespread advances in deep learning techniques, which also have been successfully applied to the vast amount of EHR data. In this paper, we review these deep EHR systems, examining architectures, technical aspects, …
Identifying patients who will be discharged within 24 hours can improve hospital resource management and quality of care. We studied this problem using eight years of Electronic Health Records (EHR) data from Stanford Hospital. We fit models to predict 24 hour discharge across the entire inpatient population. The best performing models achieved an area under the receiver-operator characteristic curve (AUROC) of 0.85 and an AUPRC of 0.53 on a held out test set. This model was also well calibrated. Finally, we analyzed the utility of this model in a decision theoretic framework to identify regio…
Electronic Health Record (EHR) data can be represented as discrete counts over a high dimensional set of possible procedures, diagnoses, and medications. Supervised topic models present an attractive option for incorporating EHR data as features into a prediction problem: given a patient's record, we estimate a set of latent factors that are predictive of the response variable. However, existing methods for supervised topic modeling struggle to balance prediction quality and coherence of the latent factors. We introduce a novel approach, the prediction-focused topic model, that uses the superv…
Managing patients with chronic diseases is a major and growing healthcare challenge in several countries. A chronic condition, such as diabetes, is an illness that lasts a long time and does not go away, and often leads to the patient's health gradually getting worse. While recent works involve raw electronic health record (EHR) from hospitals, this work uses only financial records from health plan providers (medical claims) to predict diabetes disease evolution with a self-attentive recurrent neural network. The use of financial data is due to the possibility of being an interface to internat…
The increased adoption of Electronic Health Records(EHRs) has brought changes to the way the patient care is carried out. The rich heterogeneous and temporal data space stored in EHRs can be leveraged by machine learning models to capture the underlying information and make clinically relevant predictions. This can be exploited to support public health activities such as pharmacovigilance and specifically mitigate the public health issue of Adverse Drug Events(ADEs). The aim of this article is, therefore, to investigate the various ways of handling temporal data for the purpose of detecting AD…
Peking University（北京大学）；Carnegie Mellon University（卡内基梅隆大学）；Mila - Québec AI Institute（米拉-魁北克人工智能研究所）；HEC Montréal（蒙特利尔高等商学院）
Clinical outcome prediction based on the Electronic Health Record (EHR) plays a crucial role in improving the quality of healthcare. Conventional deep sequential models fail to capture the rich temporal patterns encoded in the longand irregular clinical event sequences. We make the observation that clinical events at a long time scale exhibit strongtemporal patterns, while events within a short time period tend to be disordered co-occurrence. We thus propose differentiated mechanisms to model clinical events at different time scales. Our model learns hierarchical representationsof event sequen…
We show how to learn low-dimensional representations (embeddings) of patient visits from the corresponding electronic health record (EHR) where International Classification of Diseases (ICD) diagnosis codes are removed. We expect that these embeddings will be useful for the construction of predictive statistical models anticipated to drive personalized medicine and improve healthcare quality. These embeddings are learned using a deep neural network trained to predict ICD diagnosis categories. We show that our embeddings capture relevant clinical informations and can be used directly as input t…
Federated learning opens a number of research opportunities due to its high communication efficiency in distributed training problems within a star network. In this paper, we focus on improving the communication efficiency for fully decentralized federated learning over a graph, where the algorithm performs local updates for several iterations and then enables communications among the nodes. In such a way, the communication rounds of exchanging the common interest of parameters can be saved significantly without loss of optimality of the solutions. Multiple numerical simulations based on large…
Widespread adoption of electronic health records (EHRs) has fueled the development of using machine learning to build prediction models for various clinical outcomes. This process is often constrained by having a relatively small number of patient records for training the model. We demonstrate that using patient representation schemes inspired from techniques in natural language processing can increase the accuracy of clinical prediction models by transferring information learned from the entire patient population to the task of training a specific model, where only a subset of the population …
Georgia Institute of Technology（佐治亚理工学院）；Google Brain（谷歌大脑）；IBM Research（IBM研究院）；Sutter Health（萨特医疗）；HINT Consultants（HINT咨询公司）
Deep learning models exhibit state-of-the-art performance for many predictive healthcare tasks using electronic health records (EHR) data, but these models typically require training data volume that exceeds the capacity of most healthcare systems. External resources such as medical ontologies are used to bridge the data volume constraint, but this approach is often not directly applicable or useful because of inconsistencies with terminology. To solve the data insufficiency challenge, we leverage the inherent multilevel structure of EHR data and, in particular, the encoded relationships among…
We study the problem of detecting adverse drug events in electronic healthcare records. The challenge in this work is to aggregate heterogeneous data types involving diagnosis codes, drug codes, as well as lab measurements. An earlier framework proposed for the same problem demonstrated promising predictive performance for the random forest classifier by using only lab measurements as data features. We extend this framework, by additionally including diagnosis and drug prescription codes, concurrently. In addition, we employ a recursive feature selection mechanism on top, that extracts the top…
University of Wisconsin - Madison（威斯康星大学麦迪逊分校）；Marshfield Clinic Research Foundation（马什菲尔德诊所研究基金会）
Accurate estimation of healthcare costs is crucial for healthcare systems to plan and effectively negotiate with insurance companies regarding the coverage of patient-care costs. Greater accuracy in estimating healthcare costs would provide mutual benefit for both health systems and the insurers that support these systems by better aligning payment models with patient-care costs. This study presents the results of a generalizable machine learning approach to predicting medical events built from 40 years of data from >860,000 patients pertaining to >6,700 prescription medications, courtesy of M…
Adherence can be defined as "the extent to which patients take their medications as prescribed by their healthcare providers"[Osterberg and Blaschke, 2005]. World Health Organization's reports point out that, in developed countries, only about 50% of patients with chronic diseases correctly follow their treatments. This severely compromises the efficiency of long-term therapy and increases the cost of health services. We propose in this paper different models of patient drug consumption in breast cancer treatments. The aim of these different approaches is to predict medication non-adherence wh…
