Contextual Phenotyping of Pediatric Sepsis Cohort Using Large Language Models

2025-05-14

AI Summary

This study addresses the challenge of phenotyping pediatric sepsis patients in resource-limited settings, where high-dimensional, heterogeneous clinical data impede conventional clustering approaches and semantic interpretation. We propose the first framework integrating large language models (LLMs) into hybrid clinical data clustering. By leveraging text serialization and task-oriented prompt engineering, our method jointly encodes nutritional, clinical, and socioeconomic features to enable context-aware, interpretable subgroup identification. We evaluate quantized LLaMA-3.1-8B, LoRA-finetuned DeepSeek-R1-Distill-Llama-8B, and Stella-En-400M-V5 embeddings—each coupled with K-means—and benchmark against a UMAP+FAMD+K-medoids baseline. Stella-En-400M-V5 achieves the highest silhouette coefficient (0.86), while LLaMA-3.1 excels in multi-cluster scenarios by precisely distinguishing clinically distinct subphenotypes. Our approach significantly enhances both clustering quality and clinical interpretability, advancing data-driven sepsis phenotyping in low-resource contexts.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Clustering patient subgroups is essential for personalized care and efficient resource use. Traditional clustering methods struggle with high-dimensional, heterogeneous healthcare data and lack contextual understanding. This study evaluates Large Language Model (LLM) based clustering against classical methods using a pediatric sepsis dataset from a low-income country (LIC), containing 2,686 records with 28 numerical and 119 categorical variables. Patient records were serialized into text with and without a clustering objective. Embeddings were generated using quantized LLAMA 3.1 8B, DeepSeek-R1-Distill-Llama-8B with low-rank adaptation(LoRA), and Stella-En-400M-V5 models. K-means clustering was applied to these embeddings. Classical comparisons included K-Medoids clustering on UMAP and FAMD-reduced mixed data. Silhouette scores and statistical tests evaluated cluster quality and distinctiveness. Stella-En-400M-V5 achieved the highest Silhouette Score (0.86). LLAMA 3.1 8B with the clustering objective performed better with higher number of clusters, identifying subgroups with distinct nutritional, clinical, and socioeconomic profiles. LLM-based methods outperformed classical techniques by capturing richer context and prioritizing key features. These results highlight potential of LLMs for contextual phenotyping and informed decision-making in resource-limited settings.

Problem

• Clustering pediatric sepsis patients for personalized care
• Overcoming limitations of traditional clustering in healthcare data
• Evaluating LLM-based methods in low-income country settings

Innovation

• Uses LLM-based clustering for pediatric sepsis phenotyping
• Generates embeddings with quantized LLAMA and DeepSeek models
• Outperforms classical methods with higher Silhouette Scores