Reproducible Benchmarking for Lung Nodule Detection and Malignancy Classification Across Multiple Low-Dose CT Datasets

TL;DR

This paper introduces a reproducible, multi-dataset benchmark for lung nodule detection and malignancy classification in low-dose CT scans, highlighting dataset influence on AI model performance and promoting transparent evaluation.

Contribution

It establishes a public benchmark across multiple datasets and evaluates various training strategies, emphasizing the importance of dataset characteristics in AI performance for lung cancer detection.

Findings

Detection models trained on clinically curated datasets perform better externally.

Pretraining strategies like Strategic Warm-Start improve malignancy classification.

Dataset heterogeneity significantly impacts AI model generalizability.

Abstract

Evaluation of artificial intelligence (AI) models for low-dose CT lung cancer screening is limited by heterogeneous datasets, annotation standards, and evaluation protocols, making performance difficult to compare and translate across clinical settings. We establish a public, reproducible multi-dataset benchmark for lung nodule detection and nodule-level cancer classification and quantify cross-dataset generalizability. Using the Duke Lung Cancer Screening (DLCS) dataset as a clinically curated development set, we evaluate performance across LUNA16/LIDC-IDRI, NLST-3D, and LUNA25. Detection models trained on DLCS and LUNA16 were evaluated externally on NLST-3D using free-response ROC analysis. For malignancy classification, we compared five strategies: randomly initialized ResNet50, Models Genesis, Med3D, a Foundation Model for Cancer Biomarkers, and a Strategic Warm-Start (ResNet50-SWS) approach pretrained using detection-derived candidate patches stratified by confidence. Performance was summarized using AUC with 95% confidence intervals and DeLong tests. Detection performance varied substantially by training dataset, with DLCS-trained models outperforming LUNA16-trained models on external NLST-3D evaluation (sensitivity at 2 false positives per scan: 0.72 vs. 0.64; p < 0.001). For malignancy classification, ResNet50-SWS achieved AUCs of 0.71 (DLCS), 0.90 (LUNA16), 0.81 (NLST-3D), and 0.80 (LUNA25), consistently matching or exceeding alternative pretraining strategies. These results demonstrate that dataset characteristics strongly influence lung cancer AI performance and highlight the need for transparent, multi-dataset benchmarking.