Papers/2608.26109
🧪 Test?View on arXiv

Standalone LLM and a Pre-specified Agentic Pipeline for Explaining ICU Mortality Predictions: a Feasibility Study on the eICU Demo Dataset

Not provided in the abstract

explainabilityhealthcare AImortality predictionagentic pipelines
2608.26109
Builder Relevance
70%
1h ago

Abstract

This study explores the feasibility of using large language models and agentic pipelines to enhance the explanation of ICU mortality predictions.

Reality Card

Core Claim

The agentic pipeline improves safety-relevant grounding and patient-specific detail in ICU mortality predictions compared to standalone LLMs.

Method / Result

The agentic pipeline achieved higher guideline grounding (0.762) compared to the standalone LLM (0.143).

Limitations

The study's findings may be limited by the specific dataset used (eICU Demo Dataset) and the need for attribution-based checks before clinical application.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.
← Back to all papers