Narrative Feature or Structured Feature? A Study of Large Language Models to Identify Cancer Patients at Risk of Heart Failure
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Identifying cancer patients at risk of heart failure (HF) is crucial. Large language models (LLMs) analyzing electronic health records show promise in predicting HF risk, improving cancer patient outcomes.
Area Of Science
- Oncology
- Cardiology
- Artificial Intelligence
Background
- Cancer therapies can cause cardiotoxicity, affecting patient survival.
- Early identification of heart failure (HF) risk in cancer patients is vital for improved outcomes.
Purpose Of The Study
- To evaluate machine learning (ML) models for predicting HF risk in cancer patients.
- To explore the utility of large language models (LLMs) with novel narrative features from electronic health records (EHRs).
Main Methods
- Utilized EHR data from 12,806 cancer patients (lung, breast, colorectal) at the University of Florida Health.
- Compared traditional ML, Time-Aware long short-term memory (T-LSTM), and LLMs (GatorTron-3.9B).
- Incorporated novel narrative features derived from structured medical codes.
Main Results
- 1,602 patients developed HF post-cancer diagnosis.
- The LLM (GatorTron-3.9B) achieved superior F1 scores, outperforming SVM by 39%, T-LSTM by 7%, and BERT by 5.6%.
- Narrative features significantly enhanced model performance and feature density.
Conclusions
- LLMs demonstrate significant potential in predicting HF risk among cancer patients.
- Integrating narrative features from EHRs improves the accuracy of HF risk prediction models.
- This approach can enhance the safety and outcomes of cancer treatment.
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
Related Concept Videos
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
Genome-wide association studies or GWAS are used to identify whether common SNPs are associated with certain diseases. Suppose specific SNPs are more frequently observed in individuals with a particular disease than those without the disease. In that case, those SNPs are said to be associated with the disease. Chi-square analysis is performed to check the probability of the allele likely to be associated with the disease.
GWAS does not require the identification of the target gene involved in...
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...