LabTOP: A Unified Model for Lab Test Outcome Prediction on Electronic Health Records

2025-02-20

AI Summary

This study addresses the clinical challenges of delayed laboratory test result availability and redundant blood sampling in electronic health records (EHRs), proposing the first end-to-end continuous-value prediction framework for the full spectrum of laboratory biomarkers. Methodologically, it integrates self-supervised language modeling over EHR clinical notes, temporal-aware encoding of longitudinal lab features, and a multi-task continuous regression head—thereby overcoming limitations of prior discrete classification or subset-specific prediction approaches. Its key contribution lies in enabling unified, cross-test continuous numerical forecasting, markedly improving model generalizability and clinical deployability. Evaluated on three public EHR datasets, the framework reduces mean absolute prediction error by 23.6% relative to state-of-the-art machine learning and large language model baselines. Ablation studies confirm the substantial and complementary contributions of each component.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Natural Language Processing: (Large) Language Models
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Lab tests are fundamental for diagnosing diseases and monitoring patient conditions. However, frequent testing can be burdensome for patients, and test results may not always be immediately available. To address these challenges, we propose LabTOP, a unified model that predicts lab test outcomes by leveraging a language modeling approach on EHR data. Unlike conventional methods that estimate only a subset of lab tests or classify discrete value ranges, LabTOP performs continuous numerical predictions for a diverse range of lab items. We evaluate LabTOP on three publicly available EHR datasets and demonstrate that it outperforms existing methods, including traditional machine learning models and state-of-the-art large language models. We also conduct extensive ablation studies to confirm the effectiveness of our design choices. We believe that LabTOP will serve as an accurate and generalizable framework for lab test outcome prediction, with potential applications in clinical decision support and early detection of critical conditions.

Problem

• Predicts lab test outcomes using EHR data
• Performs continuous numerical predictions for diverse lab items
• Outperforms existing methods in accuracy and generalizability

Innovation

• Language modeling on EHR
• Continuous numerical predictions
• Outperforms traditional machine learning