Yasha Wang
YOU?
Author Swipe
View article: Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categories
Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categories Open
View article: Privacy-Preserving Federated Learning Framework for Multi-Source Electronic Health Records Prognosis Prediction
Privacy-Preserving Federated Learning Framework for Multi-Source Electronic Health Records Prognosis Prediction Open
Secure and privacy-preserving health status representation learning has become a critical challenge in clinical prediction systems. While deep learning models require substantial high-quality data for training, electronic health records ar…
View article: LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models
LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models Open
Natural Language to SQL (NL2SQL) has emerged as a critical task for enabling seamless interaction with databases. Recent advancements in Large Language Models (LLMs) have demonstrated remarkable performance in this domain. However, existin…
View article: Protocol for processing multivariate time-series electronic health records of COVID-19 patients
Protocol for processing multivariate time-series electronic health records of COVID-19 patients Open
The lack of standardized techniques for processing complex health data from COVID-19 patients hinders the development of accurate predictive models in healthcare. To address this, we present a protocol for utilizing real-world multivariate…
View article: TC–RAG: Turing–Complete RAG’s Case study on Medical LLM Systems
TC–RAG: Turing–Complete RAG’s Case study on Medical LLM Systems Open
View article: Exploring the Relationship between Dietary Intake and Clinical Outcomes in Peritoneal Dialysis Patients Stratified by Serum Albumin Levels: A 12-Year Follow-Up Using Fine-Grained Electronic Medical Records Data
Exploring the Relationship between Dietary Intake and Clinical Outcomes in Peritoneal Dialysis Patients Stratified by Serum Albumin Levels: A 12-Year Follow-Up Using Fine-Grained Electronic Medical Records Data Open
End-stage renal disease (ESRD) significantly impacts patients’ quality of life and poses substantial socioeconomic burdens. Dietary interventions are crucial for managing ESRD, yet high-quality evidence and analysis specifically linking di…
View article: GeoEdit: Geometric Knowledge Editing for Large Language Models
GeoEdit: Geometric Knowledge Editing for Large Language Models Open
View article: Enhancing Topic Interpretability for Neural Topic Modeling through Topic-wise Contrastive Learning
Enhancing Topic Interpretability for Neural Topic Modeling through Topic-wise Contrastive Learning Open
Data mining and knowledge discovery are essential aspects of extracting valuable insights from vast datasets. Neural topic models (NTMs) have emerged as a valuable unsupervised tool in this field. However, the predominant objective in NTMs…
View article: Influence of Subjective Factors on Window Use in Maternity Hospitals in Spring
Influence of Subjective Factors on Window Use in Maternity Hospitals in Spring Open
Poor indoor air quality in maternity hospitals can spread respiratory diseases; however, limited research exists on modifiable factors like occupant behavior. This study explores subjective drivers of window-opening in maternity wards, usi…
View article: RAGraph: A General Retrieval-Augmented Graph Learning Framework
RAGraph: A General Retrieval-Augmented Graph Learning Framework Open
Graph Neural Networks (GNNs) have become essential in interpreting relational data across various domains, yet, they often struggle to generalize to unseen graph data that differs markedly from training instances. In this paper, we introdu…
View article: ColaCare: Enhancing Electronic Health Record Modeling through Large Language Model-Driven Multi-Agent Collaboration
ColaCare: Enhancing Electronic Health Record Modeling through Large Language Model-Driven Multi-Agent Collaboration Open
We introduce ColaCare, a framework that enhances Electronic Health Record (EHR) modeling through multi-agent collaboration driven by Large Language Models (LLMs). Our approach seamlessly integrates domain-specific expert models with LLMs t…
View article: Protocol to process follow-up electronic medical records of peritoneal dialysis patients to train AI models
Protocol to process follow-up electronic medical records of peritoneal dialysis patients to train AI models Open
View article: TC-RAG:Turing-Complete RAG's Case study on Medical LLM Systems
TC-RAG:Turing-Complete RAG's Case study on Medical LLM Systems Open
In the pursuit of enhancing domain-specific Large Language Models (LLMs), Retrieval-Augmented Generation (RAG) emerges as a promising solution to mitigate issues such as hallucinations, outdated knowledge, and limited expertise in highly s…
View article: SMART: Towards Pre-trained Missing-Aware Model for Patient Health Status Prediction
SMART: Towards Pre-trained Missing-Aware Model for Patient Health Status Prediction Open
Electronic health record (EHR) data has emerged as a valuable resource for analyzing patient health status. However, the prevalence of missing data in EHR poses significant challenges to existing methods, leading to spurious correlations a…
View article: LoRA Dropout as a Sparsity Regularizer for Overfitting Control
LoRA Dropout as a Sparsity Regularizer for Overfitting Control Open
Parameter-efficient fine-tuning methods, represented by LoRA, play an essential role in adapting large-scale pre-trained models to downstream tasks. However, fine-tuning LoRA-series models also faces the risk of overfitting on the training…
View article: Exploring the Relationship Between Dietary Intake and Clinical Outcomes in Peritoneal Dialysis Patients
Exploring the Relationship Between Dietary Intake and Clinical Outcomes in Peritoneal Dialysis Patients Open
This study investigates the relationship between dietary intake and mortality risk among patients with End-Stage Renal Disease (ESRD), a population for whom nutritional management is crucial yet challenging due to the disease's complexity …
View article: LightM-UNet: Mamba Assists in Lightweight UNet for Medical Image Segmentation
LightM-UNet: Mamba Assists in Lightweight UNet for Medical Image Segmentation Open
UNet and its variants have been widely used in medical image segmentation. However, these models, especially those based on Transformer architectures, pose challenges due to their large number of parameters and computational loads, making …
View article: A comprehensive benchmark for COVID-19 predictive modeling using electronic health records in intensive care
A comprehensive benchmark for COVID-19 predictive modeling using electronic health records in intensive care Open
The COVID-19 pandemic highlighted the need for predictive deep-learning models in health care. However, practical prediction task design, fair comparison, and model selection for clinical applications remain a challenge. To address this, w…
View article: Learnable Prompt as Pseudo-Imputation: Rethinking the Necessity of Traditional EHR Data Imputation in Downstream Clinical Prediction
Learnable Prompt as Pseudo-Imputation: Rethinking the Necessity of Traditional EHR Data Imputation in Downstream Clinical Prediction Open
Analyzing the health status of patients based on Electronic Health Records (EHR) is a fundamental research problem in medical informatics. The presence of extensive missing values in EHR makes it challenging for deep neural networks (DNNs)…
View article: Imputation with Inter-Series Information from Prototypes for Irregular Sampled Time Series
Imputation with Inter-Series Information from Prototypes for Irregular Sampled Time Series Open
Irregularly sampled time series are ubiquitous, presenting significant challenges for analysis due to missing values. Despite existing methods address imputation, they predominantly focus on leveraging intra-series information, neglecting …
View article: Prediction model for elevated intraocular pressure risk after silicone oil filling based on clinical features
Prediction model for elevated intraocular pressure risk after silicone oil filling based on clinical features Open
Background To evaluate risk factors and further develop prediction models for intraocular pressure elevation (IOP) after vitreoretinal surgery with silicone oil tamponade to support clinical management. Methods A retrospective study analyz…
View article: Learning the Dynamic Correlations and Mitigating Noise by Hierarchical Convolution for Long-term Sequence Forecasting
Learning the Dynamic Correlations and Mitigating Noise by Hierarchical Convolution for Long-term Sequence Forecasting Open
Deep learning algorithms, especially Transformer-based models, have achieved significant performance by capturing long-range dependencies and historical information. However, the power of convolution has not been fully investigated. Moreov…
View article: Predict and Interpret Health Risk using EHR through Typical Patients
Predict and Interpret Health Risk using EHR through Typical Patients Open
Predicting health risks from electronic health records (EHR) is a topic of recent interest. Deep learning models have achieved success by modeling temporal and feature interaction. However, these methods learn insufficient representations …
View article: Mortality prediction with adaptive feature importance recalibration for peritoneal dialysis patients
Mortality prediction with adaptive feature importance recalibration for peritoneal dialysis patients Open
View article: The Combined Effects of Hourly Multi-Pollutant on the Risk of Ambulance Emergency Calls: A Seven-Year Time Series Study
The Combined Effects of Hourly Multi-Pollutant on the Risk of Ambulance Emergency Calls: A Seven-Year Time Series Study Open
Background: Ambulance emergency calls (AECs) are seen as a more suitable metric for syndromic surveillance due to their heightened sensitivity in reflecting the health impacts of air pollutants. Limited evidence has emphasized the combined…
View article: Domain-invariant Clinical Representation Learning by Bridging Data Distribution Shift across EMR Datasets
Domain-invariant Clinical Representation Learning by Bridging Data Distribution Shift across EMR Datasets Open
Emerging diseases present challenges in symptom recognition and timely clinical intervention due to limited available information. An effective prognostic model could assist physicians in making accurate diagnoses and designing personalize…
View article: A ModelOps-based Framework for Intelligent Medical Knowledge Extraction
A ModelOps-based Framework for Intelligent Medical Knowledge Extraction Open
Extracting medical knowledge from healthcare texts enhances downstream tasks like medical knowledge graph construction and clinical decision-making. However, the construction and application of knowledge extraction models lack automation, …
View article: VecoCare: Visit Sequences-Clinical Notes Joint Learning for Diagnosis Prediction in Healthcare Data
VecoCare: Visit Sequences-Clinical Notes Joint Learning for Diagnosis Prediction in Healthcare Data Open
Due to the insufficiency of electronic health records (EHR) data utilized in practical diagnosis prediction scenarios, most works are devoted to learning powerful patient representations either from structured EHR data (e.g., temporal medi…
View article: Risk factors for intraocular pressure rise after pars plana vitrectomy with silicone oil tamponade
Risk factors for intraocular pressure rise after pars plana vitrectomy with silicone oil tamponade Open
Purpose: To evaluate risk factors for intraocular pressure elevationafter vitreoretinal surgery with silicone oil tamponade and further develop predictive models to support clinical management. Methods: A retrospectively analyzed 824 eyes …
View article: KerPrint: Local-Global Knowledge Graph Enhanced Diagnosis Prediction for Retrospective and Prospective Interpretations
KerPrint: Local-Global Knowledge Graph Enhanced Diagnosis Prediction for Retrospective and Prospective Interpretations Open
While recent developments of deep learning models have led to record-breaking achievements in many areas, the lack of sufficient interpretation remains a problem for many specific applications, such as the diagnosis prediction task in heal…