MIMIC-III, a freely accessible critical care database
PulseAugur coverage of MIMIC-III, a freely accessible critical care database — every cluster mentioning MIMIC-III, a freely accessible critical care database across labs, papers, and developer communities, ranked by signal.
5 day(s) with sentiment data
-
LLM framework reconstructs patient mental health journeys from EHRs
Researchers have developed CliniCIRCA, a novel framework utilizing large language models to reconstruct longitudinal patient journeys from unstructured electronic health record (EHR) narratives. This system is designed …
-
New framework LearnActCoder improves clinical coding accuracy by learning from errors
Researchers have developed LearnActCoder, a framework designed to improve the accuracy of clinical coding agents by learning from past errors. This system creates a structured Mistake Knowledge Database (MistakeKDB) tha…
-
New hybrid AI framework improves personalized blood pressure estimation
Researchers have developed a new hybrid framework for estimating blood pressure using photoplethysmography (PPG) signals. This approach combines a convolutional neural network (CNN) with a morphology-prior branch to cap…
-
New AI framework improves hospital discharge summaries with evidence links
Researchers have developed a new framework for generating hospital discharge summaries using abstract meaning representation and deep learning. This evidence-driven approach prioritizes provenance by linking each summar…
-
New method uses medical ontologies for clinical AI generalization
Researchers have developed UdonCare, a novel method to improve domain generalization in clinical predictive healthcare. This approach leverages medical ontologies to dynamically partition patients into latent domains, a…
-
AI model predicts in-hospital stroke risk using PPG data
Researchers have developed a method to classify in-hospital stroke risk states using photoplethysmography (PPG) derived hemodynamic features. By analyzing continuous monitoring data from patients who experienced stroke …
-
New REFINE framework personalizes medical concept representation from EHRs
Researchers have developed REFINE, a novel framework designed to create personalized medical concept representations from electronic health records (EHRs). This approach refines text-attributed knowledge graphs (TKGs) b…
-
New ReTA framework dynamically augments EHR graphs with external knowledge
Researchers have developed ReTA, a novel framework that uses reinforcement learning to dynamically augment electronic health record (EHR) graphs with external knowledge graphs. This approach allows for context-aware kno…
-
New CAIR framework improves physiological time-series imputation
Researchers have developed a new two-stage framework called Curriculum-Aware Interpolate-then-Refine (CAIR) for imputing physiological time-series data, such as blood pressure and glucose levels. This method addresses l…
-
Quantum generative model TabularQGAN tackles heterogeneous tabular data synthesis
Researchers have developed TabularQGAN, a novel quantum generative model designed to synthesize tabular data, addressing a gap in existing quantum models that are typically limited to homogeneous data. This new architec…
-
New benchmark CoMedBench evaluates synthetic medical data utility
Researchers have introduced CoMedBench, a new benchmark designed to evaluate the fidelity and utility of synthetic medical data. This benchmark aims to address the challenges of using real patient data due to privacy re…
-
New framework ReCoGen generates time-series data from multimodal conditions
Researchers have developed ReCoGen, a novel two-stage framework designed to generate continuous physiological time-series data, particularly when faced with missing or irregularly sampled information. The first stage in…
-
LLM summaries enhance in-hospital mortality prediction by reorganizing clinical data
Researchers have developed a multi-representational framework that fuses LLM-generated expert summaries of ICU notes with physiological data to improve in-hospital mortality prediction. This approach significantly enhan…
-
New audit framework unmasks harmful "toxic mimicry" in medical AI
Researchers have developed a new framework called Counterfactual Clinical Audit (CCA) to identify "Toxic Mimicry" in medical offline reinforcement learning (RL) agents. This failure mode occurs when agents replicate har…
-
ReCoGen framework generates physiological time-series data from multimodal conditions
Researchers have developed ReCoGen, a novel two-stage framework designed to generate continuous physiological time-series data, particularly useful when critical signals are missing. The first stage involves training ma…
-
New GARLIC model advances interpretable AI for ICU data
Researchers have developed GARLIC, a novel neural network architecture designed to improve the accuracy and interpretability of predictive models for intensive care unit (ICU) data. This model addresses challenges like …
-
New method for causal falsification of digital twins proposed
A new research paper proposes a method for causally falsifying digital twins, which are simulation models used to predict real-world processes. The authors frame the problem as a causal inference challenge, defining wha…
-
New framework offers interpretable AI for sepsis prediction
Researchers have developed a novel framework for modeling sepsis using temporal electronic health record (EHR) data. This approach prioritizes interpretability by design, representing data relationally and then proposit…
-
New models improve causal inference for longitudinal data
Researchers have developed two new models, CSSD and CSSPD, to improve causal inference from longitudinal observational data, a crucial task for clinical decision support. These models address a fundamental tension in ex…
-
New Graph Transformer Model Enhances EHR Data Analysis for Clinical Predictions
Researchers have developed MiGHT-EHR, a novel Multi-task Graph Transformer designed to process heterogeneous temporal Electronic Health Records (EHRs). This method constructs a graph where nodes represent clinical entit…