Advances in Clinical Text Analysis and Disease Risk Prediction

The field of clinical text analysis and disease risk prediction is rapidly evolving, with a focus on developing cost-effective and scalable solutions for extracting relevant information from clinical notes and predicting disease risk. Recent developments have highlighted the potential of large language models and knowledge graphs in improving the accuracy and efficiency of clinical text analysis. Notably, the integration of knowledge graphs and Bayesian networks has shown promise in explainable disease risk prediction, while the use of diffusion models has improved the interpretability of knee osteoarthritis progression risk estimation. Furthermore, advances in clinical knowledge distillation have enabled the development of more accurate and efficient disease prediction models. Some noteworthy papers in this area include:

  • A Large Language Model Based Pipeline for Review of Systems Entity Recognition from Clinical Notes, which demonstrated the effectiveness of open-source large language models in extracting Review of Systems entities from clinical notes.
  • CKD-EHR: Clinical Knowledge Distillation for Electronic Health Records, which proposed a framework for efficient and accurate disease risk prediction through knowledge distillation techniques.

Sources

A Large Language Model Based Pipeline for Review of Systems Entity Recognition from Clinical Notes

Scalable Medication Extraction and Discontinuation Identification from Electronic Health Records Using Large Language Models

Converting Annotated Clinical Cases into Structured Case Report Forms

Integrating Knowledge Graphs and Bayesian Networks: A Hybrid Approach for Explainable Disease Risk Prediction

Risk Estimation of Knee Osteoarthritis Progression via Predictive Multi-task Modelling from Efficient Diffusion Model using X-ray Images

Adverse Event Extraction from Discharge Summaries: A New Dataset, Annotation Scheme, and Initial Findings

CKD-EHR:Clinical Knowledge Distillation for Electronic Health Records

Built with on top of