Extracting Post-Acute Sequelae of SARS-CoV-2 Infection Symptoms from Clinical Notes via Hybrid Natural Language Processing

TL;DR

This paper presents a hybrid NLP pipeline combining rule-based and BERT models to extract and detect PASC symptoms from clinical notes, improving diagnosis accuracy and efficiency across multiple health systems.

Contribution

We developed a novel hybrid NLP approach with a comprehensive PASC lexicon, validated across multiple sites, demonstrating high accuracy and processing speed for symptom extraction from clinical notes.

Findings

Achieved an average F1 score of 0.82 internally and 0.76 externally.

Processed notes in approximately 2.45 seconds each.

Showed strong correlation ($\rho > 0.83$) between model mentions and clinical diagnoses.

Abstract

Accurately and efficiently diagnosing Post-Acute Sequelae of COVID-19 (PASC) remains challenging due to its myriad symptoms that evolve over long- and variable-time intervals. To address this issue, we developed a hybrid natural language processing pipeline that integrates rule-based named entity recognition with BERT-based assertion detection modules for PASC-symptom extraction and assertion detection from clinical notes. We developed a comprehensive PASC lexicon with clinical specialists. From 11 health systems of the RECOVER initiative network across the U.S., we curated 160 intake progress notes for model development and evaluation, and collected 47,654 progress notes for a population-level prevalence study. We achieved an average F1 score of 0.82 in one-site internal validation and 0.76 in 10-site external validation for assertion detection. Our pipeline processed each note at $2.448\pm 0.812$ seconds on average. Spearman correlation tests showed $\rho >0.83$ for positive mentions and $\rho >0.72$ for negative ones, both with $P <0.0001$. These demonstrate the effectiveness and efficiency of our models and their potential for improving PASC diagnosis.