AttDMM: An Attentive Deep Markov Model for Risk Scoring in Intensive Care Units

Clinical practice in intensive care units (ICUs) requires early warnings when a patient's condition is about to deteriorate so that preventive measures can be undertaken. To this end, prediction algorithms have been developed that estimate the risk of mortality in ICUs. In this work, we propose a novel generative deep probabilistic model for real-time risk scoring in ICUs. Specifically, we develop an attentive deep Markov model called AttDMM. To the best of our knowledge, AttDMM is the first ICU prediction model that jointly learns both long-term disease dynamics (via attention) and different disease states in health trajectory (via a latent variable model). Our evaluations were based on an established baseline dataset (MIMIC-III) with 53,423 ICU stays. The results confirm that compared to state-of-the-art baselines, our AttDMM was superior: AttDMM achieved an area under the receiver operating characteristic curve (AUROC) of 0.876, which yielded an improvement over the state-of-the-art method by 2.2%. In addition, the risk score from the AttDMM provided warnings several hours earlier. Thereby, our model shows a path towards identifying patients at risk so that health practitioners can intervene early and save patient lives.

[1]  Li Li,et al.  Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records , 2016, Scientific Reports.

[2]  Anna Maria Paganoni,et al.  Multivariate Hidden Markov Models for disease progression , 2020, Stat. Anal. Data Min..

[3]  Jimeng Sun,et al.  RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism , 2016, NIPS.

[4]  E. Seow,et al.  Outcomes of direct and indirect medical intensive care unit admissions from the emergency department of an acute care hospital: a retrospective cohort study , 2014, BMJ Open.

[5]  Ahmed M. Alaa,et al.  Personalized Risk Scoring for Critical Care Prognosis Using Mixtures of Gaussian Processes , 2016, IEEE Transactions on Biomedical Engineering.

[6]  Buyue Qian,et al.  INPREM: An Interpretable and Trustworthy Predictive Model for Healthcare , 2020, KDD.

[7]  E. DeLong,et al.  Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. , 1988, Biometrics.

[8]  Peter Szolovits,et al.  MIMIC-III, a freely accessible critical care database , 2016, Scientific Data.

[9]  J. Vincent,et al.  Serial evaluation of the SOFA score to predict outcome in critically ill patients. , 2001, JAMA.

[10]  Jun S. Kim,et al.  An attention based deep learning model of clinical events in the intensive care unit , 2019, PloS one.

[11]  N. Bartolomeo,et al.  Progression of liver cirrhosis to HCC: an application of hidden Markov model , 2011, BMC medical research methodology.

[12]  Dejia Shi,et al.  Using a LSTM-RNN Based Deep Learning Framework for ICU Mortality Prediction , 2018, WISA.

[13]  Mihaela van der Schaar,et al.  ForecastICU: A Prognostic Decision Support System for Timely Prediction of Intensive Care Unit Admission , 2016, ICML.

[14]  Dipankar Bandyopadhyay,et al.  Hidden Markov models for zero‐inflated Poisson counts with an application to substance use , 2011, Statistics in medicine.

[15]  T. Lim,et al.  Risk Factors for Hospital and Long-Term Mortality of Critically Ill Elderly Patients Admitted to an Intensive Care Unit , 2014, BioMed research international.

[16]  Dongdong Zhang,et al.  Identifying Sepsis Subphenotypes via Time-Aware Multi-Modal Auto-Encoder , 2020, medRxiv.

[17]  Fei Wang,et al.  Readmission prediction via deep contextual embedding of clinical concepts , 2018, PloS one.

[18]  Søren Brunak,et al.  Dynamic and explainable machine learning prediction of mortality in patients in the intensive care unit: a retrospective study of high-frequency data in electronic patient records. , 2020, The Lancet. Digital health.

[19]  D. Bouch,et al.  Severity scoring systems in the critically ill , 2008 .

[20]  Uri Shalit,et al.  Estimating individual treatment effect: generalization bounds and algorithms , 2016, ICML.

[21]  J. Nygård,et al.  Hierarchical Hidden Markov Jump Processes for Cancer Screening Modeling. , 2019, 1910.05847.

[22]  Mihaela van der Schaar,et al.  Individualized Risk Prognosis for Critical Care Patients: A Multi-task Gaussian Process Model , 2017, ArXiv.

[23]  Aram Galstyan,et al.  Multitask learning and benchmarking with clinical time series data , 2017, Scientific Data.

[24]  Edward De Brouwer,et al.  Deep Ensemble Tensor Factorization for Longitudinal Patient Trajectories Classification , 2018, ArXiv.

[25]  Wendong Ge,et al.  An Interpretable ICU Mortality Prediction Model Based on Logistic Regression and Recurrent Neural Networks with LSTM Units , 2018, AMIA.

[26]  Michael Krauthammer,et al.  Patient Similarity Analysis with Longitudinal Health Data , 2020, ArXiv.

[27]  Stefan Feuerriegel,et al.  Early Detection of User Exits from Clickstream Data: A Markov Modulated Marked Point Process Model , 2020, WWW.

[28]  Jimeng Sun,et al.  Medical Concept Representation Learning from Electronic Health Records and its Application on Heart Failure Prediction , 2016, ArXiv.

[29]  D. Teres,et al.  Assessing contemporary intensive care unit outcome: An updated Mortality Probability Admission Model (MPM0-III)* , 2007, Critical care medicine.

[30]  Hye Jin Kam,et al.  Learning representations for the early detection of sepsis with deep neural networks , 2017, Comput. Biol. Medicine.

[31]  Ping Zhang,et al.  Risk Prediction with Electronic Health Records: A Deep Learning Approach , 2016, SDM.

[32]  Peter Szolovits,et al.  A Multivariate Timeseries Modeling Approach to Severity of Illness Assessment and Forecasting in ICU with Sparse, Heterogeneous Clinical Data , 2015, AAAI.

[33]  Peter Bauer,et al.  SAPS 3—From evaluation of the patient to evaluation of the intensive care unit. Part 2: Development of a prognostic model for hospital mortality at ICU admission , 2005, Intensive Care Medicine.

[34]  Fenglong Ma,et al.  HiTANet: Hierarchical Time-Aware Attention Networks for Risk Prediction on Electronic Health Records , 2020, KDD.

[35]  Yan Liu,et al.  Recurrent Neural Networks for Multivariate Time Series with Missing Values , 2016, Scientific Reports.

[36]  Hagen Malberg,et al.  Predicting sepsis with a recurrent neural network using the MIMIC III database , 2019, Comput. Biol. Medicine.

[37]  Stefan Feuerriegel,et al.  Modeling longitudinal dynamics of comorbidities , 2021, CHIL.

[38]  Stefan Feuerriegel,et al.  Estimating Average Treatment Effects via Orthogonal Regularization , 2021, CIKM.

[39]  Stefan Feuerriegel,et al.  Sequential Deconfounding for Causal Inference with Unobserved Confounders , 2021, ArXiv.

[40]  R. Khan,et al.  Sequential Tests of Statistical Hypotheses. , 1972 .

[41]  Sean Gerrish,et al.  Black Box Variational Inference , 2013, AISTATS.

[42]  J. Schnipper,et al.  Potentially avoidable 30-day hospital readmissions in medical patients: derivation and validation of a prediction model. , 2013, JAMA internal medicine.

[43]  Yan Liu,et al.  Benchmark of Deep Learning Models on Large Healthcare MIMIC Datasets , 2017, ArXiv.

[44]  Uri Shalit,et al.  Structured Inference Networks for Nonlinear State Space Models , 2016, AAAI.

[45]  Ding Yu-Ting,et al.  Real-time risk assessment based on hidden Markov model and security configuration , 2014, 2014 International Conference on Information Science, Electronics and Electrical Engineering.

[46]  S. Lemeshow,et al.  A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. , 1993, JAMA.

[47]  Dylan S. Small,et al.  HIDDEN MARKOV MODELS FOR ALCOHOLISM TREATMENT TRIAL DATA , 2010, 1010.1410.

[48]  Yuval Shahar,et al.  Temporal Probabilistic Profiles for Sepsis Prediction in the ICU , 2019, KDD.

[49]  Stefan Feuerriegel,et al.  Estimating Risk-Adjusted Hospital Performance , 2020, 2020 IEEE International Conference on Big Data (Big Data).

[50]  Geoffrey K Lighthall,et al.  Understanding Decision Making in Critical Care , 2015, Clinical Medicine & Research.

[51]  J. Zimmerman,et al.  Acute Physiology and Chronic Health Evaluation (APACHE) IV: Hospital mortality assessment for today’s critically ill patients* , 2006, Critical care medicine.

[52]  Satya Narayan Shukla,et al.  Modeling Irregularly Sampled Clinical Time Series , 2018, ArXiv.

[53]  Mark Nunnally,et al.  ICU Admission, Discharge, and Triage Guidelines: A Framework to Enhance Clinical Operations, Development of Institutional Policies, and Further Research , 2016, Critical care medicine.

[54]  Tosha B. Wetterneck,et al.  Hospital Readmission in General Medicine Patients: A Prediction Model , 2009, Journal of General Internal Medicine.

[55]  Fenglong Ma,et al.  Dipole: Diagnosis Prediction in Healthcare via Attention-based Bidirectional Recurrent Neural Networks , 2017, KDD.

[56]  Anna Rumshisky,et al.  Unfolding physiological state: mortality modelling in intensive care units , 2014, KDD.