Improved forecasting of suicide attempts using LSGPs for patients with little data.
problem Challenges in predicting suicide attempts due to their rarity and patient heterogeneity.
method Introduced Latent Similarity Gaussian Processes (LSGPs) to capture patient heterogeneity.
result LSGPs outperform baseline models, even without kernel-design, and offer new insights into patient similarity.
Efficiently fine-tunes patient-independent seizure detection models with tensor kernel machine.
problem Improving seizure detection accuracy for wearable devices.
method Transfer learning with tensor kernel machine using canonical polyadic decomposition.
result Patient fine-tuned model achieves high performance with smaller model size.
New method measures patient similarity over time, improving disease risk prediction.
problem Chronic diseases' varying progression rates and heterogeneous clinical presentations make patient comparison difficult.
method Subsequence alignment to account for pathophysiological misalignment and varying patient presentation times.
result Subsequence alignment outperforms global alignment in predicting disease progression.
Neural network classifies liver cancer patients based on genomic data.
problem Classifying liver cancer patients into high-risk and low-risk groups.
method Data expansion using wavelet analysis, compression of wavelet coefficients, training a neural network model.
result The neural network model accurately classifies patients without survival time information.
Deep learning clusters patient time-series data for better prognosis.
problem Clustering time-series data for patient phenotyping and prognosis.
method Deep predictive clustering with novel loss functions for future outcome distribution.
result Model achieves superior clustering performance and identifies meaningful patient subgroups.
Privacy distillation lets patients control their data for medical models.
problem Patient self-censorship due to privacy concerns impairs personalized treatment accuracy.
method Privacy distillation using deep neural networks to retain model accuracy with partial patient data.
result Privacy distillation maintains model accuracy with only 3% loss and reduces health risks by 3.9%.
Model predicts treatment initiation from clinical data using patient-clinician relations.
problem Predicting treatment initiation from clinical time series data considering patient-clinician relations.
method Graph-Augmented Time-Sensitive Model using top eigenvectors of graph Laplacian.
result Relational similarity improves prediction over baselines, e.g., 5% improvement in AUPRC.
The paper presents a method to score patient engagement in care programs and predicts their response.
problem Improving health outcomes of high-need patients through better patient engagement.
method Data-driven behavioral engagement scoring pipeline for two aspects of patient engagement.
result The scoring method successfully predicts patient engagement and provides interpretable insights.
UMAP visualizes patient phenotypes from EHR data for emergency triage.
problem Interpreting high-dimensional EHR data for rapid patient triage.
method UMAP for non-linear dimensionality reduction, Gaussian mixture models for clustering.
result UMAP reveals clinically relevant patient phenotypes from EHR data.
Machine learning detects NASH patients from medical claims data.
problem Detecting undiagnosed NASH patients for screening and management.
method Gradient-boosted decision trees trained on administrative medical claims data.
result Model precision for NASH detection is significantly higher than NASH incidence.
FRESH combines patient-level and aggregate-level data for better clinical decision making.
problem Combining patient-level and aggregate-level data for clinical decision making.
method FRESH method that re-calibrates a patient-level model to match specified aggregate statistics.
result Unified data-efficient model for clinical decision making.
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.
VAE generates synthetic patient data for training deep learning models.
problem Lack of accurate patient data for training deep learning models.
method Used Variational Autoencoders (VAE) to learn latent distribution of patient features and generate new patient records.
result VAE can generate new accurate patient records based on diagnosis.
Proposes a framework for personalized treatment recommendations using observational data.
problem Estimating patient-level treatment effects from observational data.
method Integrates existing methods for learning patient-level causal models.
result Improves patient outcomes in heart failure patients with acute kidney injury.
Patient journeys are compared to find clusters of similar disease trajectories.
problem Discovering shared health outcomes among patient journeys.
method Comparing longitudinal health data to identify clusters of similar patient trajectories.
result Clusters of patient journeys with similar health outcomes can be identified.
Study proposes a model to improve patient subtyping from EHR data.
problem Challenges in subtyping temporal EHR datasets.
method Self-supervised Mamba-based model for learning EHR representations.
result Model outperforms baseline models in EHR data subtyping.
Model identifies key problems in HIV patients' records.
problem Complex and time-consuming task of identifying patient problems from electronic health records.
method Unsupervised phenotyping approach that jointly learns phenotypes from structured and unstructured data.
result Learned phenotypes and their relatedness are clinically valid and surpass existing methods.
Algorithm scores critical care patients for clinical acuity.
problem Assessing clinical acuity of critical care patients.
method Hierarchical latent class model, mixture of Gaussian Process experts, self-taught transfer learning.
result Personalized risk score outperforms existing scores.
Study predicts risk of true-lumen narrowing after ATAAD surgery using CT data.
problem Early post-surgery risk assessment for aortic dissection patients.
method Retrospective study with CT data, derived cross-sectional shapes, form factor (FF) for morphology assessment, linear discriminant analysis (LDA) for risk classification, LOPO-CV for prediction.
result Machine-learning model accurately predicts risk for all high-risk patients and low-risk patients, potentially reducing hospital visits.
Proposes a deep learning framework for evaluating patient similarities from EHRs.
problem Evaluating clinical similarities between patients for various healthcare applications.
method A deep learning framework with medical concept embedding, preserving temporal information.
result Significant improvement in patient similarity evaluation over baselines.
Method infers TCR effects on disease using patient data.
problem Estimating causal effects of TCR sequences on patient outcomes.
method Corrects for confounders using pre-selection TCR data and neural-network estimator.
result Identifies potentially therapeutic TCRs for COVID-19 severity.
A new model uses GPs and latent force models to predict patient responses to drugs.
problem Challenges in modeling short-term effects of drugs on patient physiology.
method Hybrid Gaussian process with latent force model for joint modeling of patient physiology and drug effects.
result The model accurately predicts patient responses to three common drugs, showing competitive performance.
The paper tracks patient recovery using graphs of joint movement data.
problem Tracking individual patient recovery trajectories in physical therapy.
method Bayesian learning of Random Geometric Graphs from joint movement data.
result Optimal exercise routines can be recommended based on patient recovery data.
The study uses demographical data to predict health conditions.
problem Predicting health conditions based on patient demographics and symptoms.
method Analyzing Electronic Health Records (EHR) from Brazil to identify age-related clusters of health conditions.
result Age of patients significantly influences the likelihood of certain health conditions.
Deep Learning improves end-of-life care by predicting patient mortality.
problem Misalignment between patient wishes and actual care at the end of life.
method Deep Neural Network trained on EHR data to predict mortality and identify patients in need of palliative care.
result Proactive approach to reaching out to patients in need of palliative care.
New approach learns missing data representations for better patient outcome predictions.
problem Missing data, diverse modalities, and complex relationships in EHRs.
method Representation learning using message passing.
result Competitive or superior performance in predicting patient outcomes.
Model predicts wound and episode-level readmission risk and time to re-admit.
problem Identify patients at high risk of re-admission to prevent wound recurrences and reduce healthcare costs.
method Data-driven analysis of wound care and episode-level patient data.
result Model achieves high recall and precision for predicting re-admission risk and time.
Clusters of ACS patients identified for better therapeutic stratification.
problem Data-driven classification and subtyping of ACS patients for improved treatment.
method Outcome-driven clustering using a multi-task neural network with attention.
result Seven patient clusters with distinct characteristics and risk profiles identified.
New method subtypes irregular patient data for disease progression.
problem Irregular observation patterns in patient data.
method Probabilistic model + mixture model for asynchronous trajectories.
result 13% reduction in cross-entropy error for vital signs forecasting.
Project predicts Alzheimer's progression using neural networks and novel data processing.
problem Difficulty in early identification of Alzheimer's patients.
method Used machine learning, specifically neural networks, and a novel pre-processing technique.
result Neural network model accurately predicts AD progression with high accuracy.
Study improves glaucoma prediction accuracy by aggregating clustering-based models.
problem Predicting glaucomatous visual field loss from limited patient data.
method Hierarchically aggregating clustering-based predictors to enhance prediction accuracy.
result Hierarchical aggregation of cluster-based predictors outperforms single methods.
Study found bias in drug effectiveness due to secular trend, adjusting for it was difficult.
problem Secular trend bias in drug effectiveness study.
method Built a machine learning causal inference model to identify subpopulations and adjust for bias using two methods.
result Bias remained even after adjusting for secular trend, suggesting other unmeasured factors.
Machine learning predicts trauma patient mortality risk.
problem Predicting mortality risk in trauma patients using traditional regression models.
method Transfer learning-based machine learning algorithm applied to trauma patient data.
result Machine learning model achieved similar performance to contemporary models without restrictive criteria.
T-LSTM autoencoder improves chronic kidney disease patient representation.
problem Improving latent representations from irregularly sampled clinical data.
method Time-Aware Long Short-Term Memory Autoencoder.
result Significant improvements in learnt representations on synthetic and real datasets.
The paper introduces Precision Disease Networks (PDN) for predicting medical outcomes.
problem Predicting medical outcomes for patients with diseases.
method Building patient-specific disease networks, clustering, and data visualization.
result PDN improves prediction of patient outcomes compared to standard statistical analysis.
Language models improve clinical prediction models using EHR data.
problem Limited patient data for training clinical prediction models.
method Using patient representation schemes from natural language processing.
result 3.5% mean improvement in AUROC on five prediction tasks.
TAPER learns unified patient EHR representations for healthcare tasks.
problem Irregular and multimodal data in electronic health records.
method Transformer networks and BERT for embedding structured and unstructured data.
result TAPER model outperforms on mortality, readmission, and length of stay tasks.
MedGP improves online patient health status prediction using clinical and lab covariates.
problem Real-time monitoring of hospital patients for accurate health status inference.
method Bayesian nonparametric Gaussian process regression with a sparse kernel.
result MedGP significantly improves online prediction accuracy for patient health status across different disease subgroups and studies.
Deep learning model predicts severe COVID-19 outcomes.
problem Predicting severe COVID-19 outcomes in ED patients.
method Deep feature fusion model using EHR data and CXR images.
result CO-RISK score achieved AUC of 0.95 and 0.92 for 24 and 72 hours predictions, superior to human performance.
Deep learning models interpret patient outcomes better with time aggregation.
problem Chronic ambulatory care data challenges for deep learning models.
method Time-distributed-dense layers combined with GRUs for generalization, clinical interpretation framework.
result Time-distributed-dense layers with GRUs produce the most generalizable models.
Paper presents a new method for clustering patient records using tensor decomposition.
problem Clustering high-dimensional binary data, especially in healthcare records.
method Tensor decomposition for an efficient and robust heuristic.
result Clinically meaningful results obtained on two healthcare datasets.
CLOCS uses contrastive learning to improve cardiac signal representations.
problem Lack of labelled data in cardiac signal analysis.
method Contrastive learning across space, time, and patients.
result CLOCS outperforms state-of-the-art methods and achieves strong generalization.
Enhances understanding of patient healthcare journeys using self-attention.
problem Capturing hidden dependencies in multi-level patient journey data.
method Proposes a multi-level self-attention network (MusaNet) for encoding patient journeys.
result MusaNet produces higher-quality representations than state-of-the-art methods.
Ward2ICU dataset protects patient privacy while generating synthetic ICU transitions data.
problem Protecting patient privacy while creating synthetic ICU transition data.
method Wasserstein Generative Adversarial Network (GAN) to generate synthetic data, class label balancing.
result Quality of synthetic data generation assessed through binary classification task.
Study identifies similarities in refractory epileptic patients to predict drug resistance.
problem Predict drug resistance in epileptic patients.
method Examined patient data using unsupervised and supervised algorithms to map underlying mechanisms and features contributing to drug resistance.
result Developed predictive models with accuracy of 0.83(+/-0.3) using a radial basis function kernel PCA and Gradient Boosted Decision Tree Ensemble.
Bayesian model predicts patient survival from sparse EHR data.
problem Analyzing EHR data with few samples and diverse information.
method Nonparametric probabilistic model using Bayesian trees.
result Improved survival trajectory predictions on patient data.
Model predicts missed doses and targets TB patients more effectively.
problem Improving adherence to TB treatment using digital data.
method Deep learning model trained on 17k TB patients' adherence data.
result Model predicts 21% more patients at risk of missing doses.
SplitNN enables deep learning model training without sharing patient data.
problem Training deep learning models without exposing sensitive patient data.
method Distributed deep learning method called SplitNN.
result SplitNN outperforms other distributed learning methods in performance and resource efficiency.