Re-thinking and Re-labeling LIDC-IDRI for Robust Pulmonary Cancer Prediction

TL;DR

This paper proposes a re-labeling strategy for the LIDC-IDRI lung cancer dataset using metric learning and a small pathological dataset, improving prediction robustness and addressing annotation bias.

Contribution

It introduces a novel re-labeling method for LIDC data based on metric learning and validates its effectiveness with improved model performance.

Findings

Re-labeling with metric learning improves prediction accuracy.

Adding uncertain nodule labels further enhances model performance.

Re-labeling mitigates annotation bias in lung cancer datasets.

Abstract

The LIDC-IDRI database is the most popular benchmark for lung cancer prediction. However, with subjective assessment from radiologists, nodules in LIDC may have entirely different malignancy annotations from the pathological ground truth, introducing label assignment errors and subsequent supervision bias during training. The LIDC database thus requires more objective labels for learning-based cancer prediction. Based on an extra small dataset containing 180 nodules diagnosed by pathological examination, we propose to re-label LIDC data to mitigate the effect of original annotation bias verified on this robust benchmark. We demonstrate in this paper that providing new labels by similar nodule retrieval based on metric learning would be an effective re-labeling strategy. Training on these re-labeled LIDC nodules leads to improved model performance, which is enhanced when new labels of uncertain nodules are added. We further infer that re-labeling LIDC is current an expedient way for robust lung cancer prediction while building a large pathological-proven nodule database provides the long-term solution.