Improving VTE Identification through Adaptive NLP Model Selection and Clinical Expert Rule-based Classifier from Radiology Reports

TL;DR

This paper introduces a novel combination of deep learning, adaptive NLP model selection, data augmentation, and expert rule-based classifiers to improve the accuracy of VTE detection from radiology reports, achieving over 97% accuracy and F1 scores.

Contribution

It presents a new integrated approach that combines deep learning, adaptive NLP model selection, data augmentation, and expert rules for better VTE identification from unstructured radiology reports.

Findings

Achieved 97% accuracy and 97% F1 score for DVT detection.

Achieved 98.3% accuracy and 98.4% F1 score for PE detection.

Demonstrated robustness and potential impact on VTE research.

Abstract

Rapid and accurate identification of Venous thromboembolism (VTE), a severe cardiovascular condition including deep vein thrombosis (DVT) and pulmonary embolism (PE), is important for effective treatment. Leveraging Natural Language Processing (NLP) on radiology reports, automated methods have shown promising advancements in identifying VTE events from retrospective data cohorts or aiding clinical experts in identifying VTE events from radiology reports. However, effectively training Deep Learning (DL) and the NLP models is challenging due to limited labeled medical text data, the complexity and heterogeneity of radiology reports, and data imbalance. This study proposes novel method combinations of DL methods, along with data augmentation, adaptive pre-trained NLP model selection, and a clinical expert NLP rule-based classifier, to improve the accuracy of VTE identification in unstructured (free-text) radiology reports. Our experimental results demonstrate the model's efficacy, achieving an impressive 97\% accuracy and 97\% F1 score in predicting DVT, and an outstanding 98.3\% accuracy and 98.4\% F1 score in predicting PE. These findings emphasize the model's robustness and its potential to significantly contribute to VTE research.