SpiroLLM: Finetuning Pretrained LLMs to Understand Spirogram Time Series with Clinical Validation in COPD Reporting

2025-07-21

AI Summary

Current AI models for COPD diagnosis lack interpretability, and mainstream large language models (LLMs) cannot process spirometry graphs, severely limiting clinical trustworthiness and real-world deployment. To address this, we propose the first multimodal LLM capable of comprehending pulmonary function graphs. Our method introduces a unified latent-space framework that jointly encodes morphological features from spirometric time-series curves and aligns them with numerical pulmonary function indices. Specifically, SpiroEncoder extracts temporal flow patterns, while SpiroProjector maps both curve features and clinical metrics into a shared latent space; an LLM is then fine-tuned to generate interpretable diagnostic reports. Evaluated on 234,028 subjects from UK Biobank, our model achieves an AUROC of 0.898. Crucially, it maintains 100% response rate under critical data missingness—substantially outperforming text-only baselines (13.4%).

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Multi-modal Vision
• Natural Language Processing: (Large) Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Large language models for search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Chronic Obstructive Pulmonary Disease (COPD), a major chronic respiratory disease with persistent airflow limitation, is a leading global cause of disability and mortality. Respiratory spirogram time series, routinely collected during pulmonary function tests (PFTs), play a critical role in the early detection of repsiratory diseases and in monitoring lung function over time. However, most current AI models for COPD diagnosis are limited to outputting classification results without providing a rationale for their diagnostic process, while current Large Language Models (LLMs) cannot understand spirograms yet, which severely limits their clinical trust and adoption. To tackle this challenge, we leverage a cohort of 234,028 individuals from the UK Biobank (UKB) to propose SpiroLLM, the first multimodal large language model that can understand spirogram. The model extracts morphological features from respiratory curves via a SpiroEncoder and aligns them with PFT numerical values in a unified latent space using a SpiroProjector, ultimately empowering a large language model to generate a comprehensive diagnostic report. Experimental results confirm that SpiroLLM achieved a diagnostic AUROC of 0.8980 (95% CI: 0.8820-0.9132). In a robustness test with missing core data, it maintained a 100% valid response rate, far surpassing the 13.4% of a text-only model and showcasing the superiority of its multimodal design. This work demonstrates the substantial potential of deeply fusing physiological signals with large language models, establishing a new paradigm for the next generation of interpretable and reliable clinical decision support tools.

Problem

• Enabling LLMs to understand spirogram time series for COPD diagnosis
• Providing interpretable diagnostic reports beyond simple classification
• Improving clinical trust via multimodal fusion of physiological and textual data

Innovation

• Finetunes LLMs to understand spirogram time series
• Uses SpiroEncoder to extract morphological features
• Aligns features with PFT values via SpiroProjector