Improved RNN predicts patient health from ICU data.
problem Slowed response times and missed true alarms from ICU alarms.
method Compare sliding window and recurrent predictors for ICU multivariate time series.
result RNN slightly improved for three out of four patient state-of-health targets.
RAIM models ICU patient data for better clinical decision support.
problem Challenges in analyzing high-density, heterogeneous patient monitoring data.
method RAIM integrates continuous monitoring data and discrete clinical events using an attention mechanism.
result RAIM predicts physiological decompensation and length of stay with high accuracy.
Reduces false alarms in ICU monitoring systems.
problem High false alarm rates in ICU monitoring systems.
method Multitask learning with distant supervision.
result Significant improvements over state-of-the-art baselines.
Clinical notes improve ICU patient condition prediction.
problem Predicting ICU patient outcomes and resource planning.
method Joint modeling of time series data and clinical notes.
result Significant improvement in mortality prediction, decompensation modeling, and length of stay forecasting.
Paper tackles ICU false alarms by learning features from ECG signals.
problem High rate of false alarms in ICU due to patient movements and sensor detachment.
method Unsupervised feature learning to extract and cluster high-level features from ECG signals.
result The method reduces false arrhythmia alarms using a few high-level features from a single ECG lead.
Study develops electronic phenotypes of ICU patient acuity.
problem Limited time for patient acuity assessments and imprecise clinical trajectory prediction.
method Developed electronic phenotypes using automated variable retrieval in electronic health records.
result Identified three phenotypes: persistently stable, persistently unstable, and transitioning from unstable to stable.
Machine learning predicts circulatory failure in ICU patients.
problem Limited ability of clinicians to recognize early signs of patient deterioration.
method Developed an early warning system using machine learning on ICU data.
result Predicts 90.0% of circulatory failure events with 81.8% identified more than two hours in advance.
Model predicts blood culture positivity, aiding ICU doctors.
problem Early detection of bacteria or fungi in blood cultures.
method BiLSTM network trained on ICU clinical time series data.
result 71.95% AUC under precision-recall curve.
DeepSOFA acuity score uses deep learning for ICU patient severity assessment.
problem Static severity scoring methods are time-consuming and inaccurate for ICU patients.
method Temporal measurements and interpretable deep learning models.
result DeepSOFA yields significantly more accurate predictions of in-hospital mortality.
System predicts respiratory failure up to 8 hours early.
problem Early detection of respiratory failure in ICU patients.
method Machine learning on ICU patient monitoring data.
result System outperforms traditional clinical decision-making.
Bayesian model detects internal bleeding in ICU patients.
problem Late detection of internal bleeding in ICU patients.
method Bayesian regime switching model analyzing vitals and labs.
result Probabilistic assessment of patient's physiological state.
Develops a model to predict clinical deterioration in ICU patients.
problem Predicting clinical deterioration in critically ill patients.
method Semi-Markov Switching Linear Gaussian Model (SSLGM) with censored data.
result SSLGM significantly outperforms existing risk scores.
Enhances causal estimation using unlabeled offline ICU data.
problem Assessing unmeasured physiological variables in new ICU patients.
method Three-stage approach: non-causal and causal estimators, causal filter, and prediction for new patients.
result Enhanced causal estimation for new ICU patients using offline data.
The paper develops a method to predict ICU mortality risk across diverse patient populations.
problem Improving patient survival by recognizing risky trajectories during ICU stays.
method Domain adaptation strategies to learn mortality prediction models robust to diverse ICU populations.
result The proposed model outperforms baselines, achieving AUC numbers up to 0.88 for the Cardiac ICU population.
Estimates future ICU demand using age-structured data and logistic modeling.
problem Projecting future ICU demand from low-quality data.
method Age-structured sampling, logistic fit, subnotification factor correction.
result Corrected logistic curve projects future ICU demand.
RNN predicts ICU mortality trajectories from pediatric ICU data.
problem Predicting in-ICU mortality for pediatric patients.
method Developed a recurrent neural network using EMR data.
result RNN predictions significantly outperform existing methods.
Deep learning model predicts ICU mortality by attending to missing data.
problem Handling irregular and missing data in ICU physiological time-series.
method Long Short-Term Memory architecture with attention mechanisms.
result Competitive and interpretable results on PhysioNet 2012 dataset.
Study developed phenotypes for ICU patients' brain dysfunction states.
problem Underdiagnosis of acute brain dysfunction in ICU patients.
method Created algorithms to quantify and cluster brain dysfunction states.
result Developed three phenotypes of ICU patients' brain dysfunction states.
Good predictors of ICU Mortality have the potential to identify high-risk patients earlier, improve ICU resource allocation, or create more accurate population-level risk models. Machine learning practitioners typically make choices about how to represent features in a particular model, but these choices are seldom eva…
Study identifies sepsis subpopulations for better ICU prediction.
problem Lack of stratified sepsis subpopulations in ICU EHR predictions.
method Stratified sepsis subpopulations identified using machine learning.
result Machine learning models perform better with stratified subpopulations.
Study proposes a self-correcting deep learning model for ICU patient condition prediction.
problem Challenging task of continuously monitoring high-dimensional vital signs and lab measurements in critical care.
method Utilized accumulative ICU data, self-correcting mechanism, and regularization method.
result Outperformed conventional deep learning models in predicting acute kidney injury.
Natural language processing predicts ICU readmissions with 74.8% accuracy.
problem Early detection of ICU readmissions to improve patient outcomes and reduce costs.
method Natural language processing of discharge summaries, machine learning classifiers, UMLS standardization.
result Best configuration achieved an AUC of 0.748 for predicting ICU readmissions.
Robust policies improve ICU transfer outcomes by predicting patient deterioration.
problem Higher mortality rates for unplanned ICU transfers.
method Markov Decision Process model to predict patient severity and optimize transfer policies.
result Robust policies are more aggressive in transferring patients than nominal policies, improving overall patient care.
Study describes severe dengue ICU patients in Brazil, 2012-2024.
problem Characterize severe dengue ICU patients and identify risk factors.
method Prospective study, descriptive statistics, logistic regression, machine learning.
result Advanced age, comorbidities, leukocytes, and platelets are significant risk factors for complications.
Invasive interventions did not significantly shorten ICU patients' time to death.
problem Determining if withholding or withdrawing invasive interventions affects ICU patients' time to death.
method Observational study using MIMIC-III ICU database, stratifying by severity, comparing 2002-2005 vs. 2008-2011.
result Despite a reduction in invasive interventions, time to death did not decrease.
Late fusion of clinical notes and physiological data improves ICU mortality prediction.
problem Improving ICU mortality prediction using multimodal data.
method Late fusion of clinical notes and physiological time series data with a deep learning architecture.
result Late fusion approach provides statistically significant improvement in mortality prediction performance.
Novel metrics improve machine learning models for ICU patient care.
problem Predicting vital sign trajectories for early detection of adverse events.
method Developed novel performance metrics aligned with clinical contexts, validated on simulated and real datasets, and optimized neural networks using these metrics.
result Neural networks trained with these metrics excel in predicting clinically significant events.
Study proposes managing COVID-19 without economy shutdowns.
problem Avoiding lockdowns while maintaining healthcare system capacity.
method Detailed heterogeneous epidemiological model, calibrated to data.
result Countries can avoid lockdowns if ICU beds per million > 100.
Predicts physiologically acceptable states for pediatric ICU discharge.
problem Determining physiologically stable states for pediatric ICU discharge.
method Computed PASS values from hr, sbp, and dbp measurements, compared to age-normal and polynomial regression predictions.
result RNN model predictions were more accurate than age-normal vitals.
Study discovers rules linking patient symptoms to unplanned ICU transfers.
problem Identifying patients at risk for unplanned ICU transfers.
method Mixed-integer optimization approach to learn association rules.
result Significant rules discovered for each patient subgroup.
Bayesian models improve trustworthiness in ICU by providing uncertainty.
problem Uncertainty in machine learning predictions can lead to catastrophic medical decisions.
method Bayesian Neural Network and predictive uncertainty analysis.
result Bayesian models can mitigate prediction loss and identify out-of-domain examples.
Supervised NMF improves ICU mortality risk prediction.
problem Predicting ICU mortality risk from complex, temporal data.
method Adopted Subgraph Augmented Nonnegative Matrix Factorization (SANMF) and integrated logistic regression loss function.
result Supervised SANMF outperformed conventional methods in ICU mortality risk prediction.
Paper uses RL to optimize ICU load during COVID-19.
problem Optimizing ICU load during a pandemic.
method Combines epidemic model, Bayesian inference, and RL for adaptive intervention levels.
result RL policies reduce ICU burden compared to historical interventions.
Paper proposes a new method to handle missing data in medical records using sequential variational autoencoders.
problem Missing data in medical records due to sensor off-times and uneven data collection.
method Sequential variational autoencoders (VAEs) with a new methodology called Shi-VAE.
result Shi-VAE achieves the best performance in terms of both metrics compared to state-of-the-art methods.
WRSE predicts dynamic survival distributions in ICU patients.
problem Dynamic assessment of ICU patient mortality risk.
method Non-parametric weighted-resolution ensemble model combining binary classifiers.
result Competitive results with state-of-the-art models, reducing training time.
Cancer patients admitted to ICU had improved survival over 10 years.
problem To assess changes in survival of cancer patients admitted to ICU over 10 years.
method Retrospective analysis of MIMIC-III database, adjusted for confounders using logistic regression.
result Cancer patients had significantly lower 28-day and 1-year mortality rates over 10 years.
The paper tackles ICU discharge strategies by evaluating optimal stopping scenarios.
problem Optimal stopping in ICU discharge with complex causal inference challenges.
method Generalized g-formula implementation and application to MIMIC-IV dataset.
result Improved ICU discharge strategies demonstrated with the g-formula approach.
New method predicts ICU stay for pancreatitis patients.
problem Predicting ICU stay for pancreatitis patients.
method Survival-supervised topic modeling with elastic-net regularized Cox model and anchor words.
result Our method is as accurate as best baselines but more interpretable.
RNN models perform similarly with or without extraneous variables.
problem Impact of extraneous variables on RNN performance in clinical tasks.
method Investigated the effect of extraneous input variables on RNN predictive performance using EMR and randomly drawn variables.
result Degradations in RNN's predictive performance with extraneous variables were negligible.
Deep learning predicts ICU mortality with enhanced interpretability.
problem Improving mortality prediction accuracy and clinician trust in AI.
method Trained a deep learning model on MIMIC-III to interpret nursing notes.
result Model reaches ROC of 0.8629, outperforming SAPS-II.
Deep learning models predict ICU readmission with varying accuracy.
problem Predicting ICU readmission risk using deep learning architectures.
method Several deep learning architectures including attention-based models, recurrent layers, neural ODEs, and embeddings were trained on MIMIC-III data.
result Attention-based models with neural ODEs achieved highest predictive accuracy.
DBN improves sepsis mortality risk prediction in ICU patients.
problem Early sepsis detection in ICU patients with infection.
method Data-driven, semantically enhanced dynamic Bayesian network.
result DBN AUROC 0.91, outperforming other scoring tools.
Natural language processing predicts AKI onset in ICU patients.
problem Early detection of AKI in ICU patients to improve outcomes.
method Clinical notes were processed to generate word and concept embeddings. Five classifiers and a deep learning model were used to predict AKI.
result The best model achieved an AUC of 0.779 for predicting AKI onset.
Novel IRL method identifies suboptimal medical decisions in ICU data.
problem Identifying suboptimal medical decisions in clinical settings.
method Incorporates Inverse Reinforcement Learning with a pruning step to identify and remove suboptimal actions.
result Pruning step effectively identifies clinical priorities and values from suboptimal data.
New model predicts ICU patients' stay duration efficiently.
problem Efficient ICU bed allocation under resource constraints.
method Temporal Pointwise Convolution (TPC) model combining temporal and pointwise convolutions.
result Significant performance improvements over LSTM and Transformer models.
Improved aggregation methods learn from all ICU events without preprocessing for better patient risk analysis.
problem Lack of efficient methods to dynamically assess patient status in ICU.
method Improved aggregation methods for a deep learning architecture that learns from all events without preprocessing.
result Models achieve strong performance (AUROC 0.87) in patient mortality classification.
Study predicts blood pressure response to fluid bolus therapy with high accuracy.
problem Predicting successful response to fluid bolus therapy in hypotensive ICU patients.
method Used attention-based LSTM and GRU neural networks on a large ICU database.
result Stacked LSTM with attention mechanism achieved highest accuracy of 0.852.
Study improves mortality prediction in ICU patients using feature engineering and 1D CNN.
problem Improving mortality prediction in ICU patients with high-dimensional, imbalanced, and missing data.
method Feature engineering, 1D Convolutional Neural Network (1D CNN), traditional machine learning algorithms.
result Best AUC of 0.848 achieved with 1D CNN model.