Tri-Reader: An Open-Access, Multi-Stage AI Pipeline for First-Pass Lung Nodule Annotation in Screening CT
This work addresses the need for efficient and accurate lung nodule annotation in medical imaging, which is incremental as it combines existing open-access models into a unified workflow.
The authors tackled the problem of automating lung nodule annotation in screening CT scans by developing Tri-Reader, a multi-stage AI pipeline that integrates segmentation, detection, and classification, resulting in a system that prioritizes sensitivity and reduces candidate burden for annotators.
Using multiple open-access models trained on public datasets, we developed Tri-Reader, a comprehensive, freely available pipeline that integrates lung segmentation, nodule detection, and malignancy classification into a unified tri-stage workflow. The pipeline is designed to prioritize sensitivity while reducing the candidate burden for annotators. To ensure accuracy and generalizability across diverse practices, we evaluated Tri-Reader on multiple internal and external datasets as compared with expert annotations and dataset-provided reference standards.