Machine Learning in Healthcare

A graph-embedded topic model enables characterization of diverse pain phenotypes among UK Biobank individuals

The paper introduces a graph-embedded topic model to analyze UK Biobank data, identifying pain phenotypes and revealing cardiovascular-related medication and condition patterns.

Configuring a federated network of real-world patient health data for multimodal deep learning prediction of health outcomes.

The paper presents a modular pipeline for federated health data analysis using deep learning to predict health outcomes across diseases.

Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET) Using Electronic Health Record Data

SCORNET estimates survival curves using EHR data with noisy event times and limited current status labels, achieving efficiency similar to Weibull regression with non-parametric flexibility.

Pancreatic cancer risk predicted from disease trajectories using deep learning

This paper uses deep learning to predict pancreatic cancer risk from clinical disease trajectories in two large datasets, achieving higher accuracy than models without temporal information.

Phenotypic signatures in clinical data enable systematic identification of patients for genetic testing

The paper develops a model to identify patients likely to have a genetic disease using EHR data, achieving high accuracy in predicting genetic testing needs.

Enhanced COVID-19 data for improved prediction of survival

The paper enhances a COVID-19 dataset by adding weather and research sentiment features, achieving 95% accuracy in survival prediction using XGBoost.

A multi-modal machine learning approach towards predicting patient readmission

This paper develops and compares machine learning models to predict 30-day unplanned readmission in frail elderly patients using electronic health records.

Comparative analysis of EHR-based stroke phenotyping methods, their applications, and interpretation

The paper compares machine learning algorithms for identifying acute ischemic stroke patients using EHR data, finding high accuracy with logistic regression and demonstrating generalizability.

High-throughput Phenotyping with Temporal Sequences

This study develops a high-throughput phenotyping method using temporal sequences from EHR data, achieving superior classification performance compared to conventional methods.

Using indication embeddings to represent patient health for drug safety studies

This paper develops indication embeddings to reduce health data dimensionality, improving cohort matching for drug safety studies by capturing relevant treatment decision information.

sureLDA: A Multi-Disease Automated Phenotyping Method for the Electronic Health Record

sureLDA automates phenotype labeling in EHR data using topic modeling, achieving high accuracy and robustness across various phenotypes.

Leveraging UMLS-driven NLP to enhance identification of influenza predictors derived from electronic medical record data

This study uses UMLS-driven NLP on EHR data to identify influenza predictors, finding that complex models with NLP-enhanced data outperform simpler models.

Predicting 30-day Hospital Readmissions Using Artificial Neural Networks with Medical Code Embedding

This study compares prediction models for 30-day hospital readmissions, finding that artificial neural networks with medical code embeddings improve accuracy compared to logistic regression.

Development and Validation of Phenotype Classifiers across Multiple Sites in the Observational Health Sciences and Informatics (OHDSI) Network

The paper develops and validates phenotype classifiers across multiple sites using machine learning, showing good portability within the US but limited international generalizability.

Multicenter validation of a machine learning algorithm for 48 hour all-cause mortality prediction

This paper validates a machine learning algorithm for predicting 48-hour all-cause mortality across multiple centers, showing high AUROC values compared to qSOFA and MEWS.

Tensor Factorization-based Prediction with an Application to Estimating the Risk of Chronic Diseases

This paper develops tensor factorization-based models to predict individual chronic disease risk using EHR data, achieving 60% recall in top 10 predictions.

SALTClass: classifying clinical short notes using background knowledge from unlabeled data

SALTClass improves classification of sparse clinical text by using clustering and smoothing on unlabeled data to enhance representation for supervised classification tasks.

Integrating Biomedical Research and Electronic Health Records to Create Knowledge Based Biologically Meaningful Machine Readable Embeddings

The paper integrates biomedical research with electronic health records to create biologically meaningful machine-readable embeddings for precision medicine.

Analysis and Prediction of Unplanned Intensive Care Unit Readmission using Recurrent Neural Networks with Long Short-Term Memory

This paper uses LSTM-based RNNs to predict ICU readmissions from EHR data, achieving higher sensitivity and AUC than traditional methods.

Augmented Intelligence with Natural Language Processing Applied to Electronic Health Records is Useful for Identifying Patients with Non-Alcoholic Fatty Liver Disease at Risk for Disease Progression

This study shows that natural language processing applied to electronic health records can accurately identify patients with non-alcoholic fatty liver disease and detect those at risk for disease progression.

Analyzing the heterogeneity of rule-based EHR phenotyping algorithms in CALIBER and the UK Biobank

This study analyzes 70 EHR phenotyping algorithms from CALIBER and UK Biobank to understand their heterogeneity and identify common implementation patterns for developing a minimum information standard.

Prediction performance of a cardiovascular risk assessment tool using Stanford EHR data repository

The paper develops and evaluates prediction models for cardiovascular diseases and type 2 diabetes using EHR data, showing high discrimination performance for most conditions.

High-throughput Multimodal Automated Phenotyping (MAP) with Application to PheWAS

This paper develops an automated high-throughput phenotyping method (MAP) that improves phenotype accuracy and scalability for PheWAS studies.

Hierarchical Patient-centric Caregiver Network Method for Clinical Outcomes Study

The paper models caregiver networks around individual patients to predict clinical outcomes like hospital stay and mortality.

Removing Confounding Factors Associated Weights in Deep Neural Networks Improves the Prediction Accuracy for Healthcare Applications

This paper introduces a method to remove confounding factors in deep neural networks to improve prediction accuracy in healthcare applications.

Deep Bidirectional Recurrent Neural Networks as End-To-End Models for Smoking Status Extraction from Clinical Notes in Spanish

This paper compares a deep bidirectional RNN end-to-end model with traditional classifiers for smoking status extraction from Spanish clinical notes, finding the RNN achieves high accuracy with significantly less development time.

Opportunities And Obstacles For Deep Learning In Biology And Medicine

Deep learning rivals or beats baselines across imaging, health records, genomics and drug discovery but is not yet transformative; scarce labels, privacy and opacity remain hurdles.

Latent-Based Imputation of Laboratory Measures from Electronic Health Records: Case for Complex Diseases

The paper improves imputation of missing lab measures in IBD EHR data using latent-based analysis and clustering of comorbidities.

Analysis of Risk Factor Domains in Psychosis Patient Health Records

The paper develops a topic extraction pipeline from psychiatric EHRs to support readmission risk prediction, showing initial results and future feature plans.

Flexible, Cluster-Based Analysis of the Electronic Medical Record of Sepsis with Composite Mixture Models

This paper develops a composite mixture model to analyze sepsis EMR data, identifying mortality risk trends and subgroups.