False Positive Reduction by Actively Mining Negative Samples for Pulmonary Nodule Detection in Chest Radiographs

TL;DR

This paper introduces a semi-supervised learning approach that actively mines negative samples from unlabeled chest radiographs to significantly reduce false positives in pulmonary nodule detection while preserving high sensitivity.

Contribution

The study presents a novel semi-supervised method that uses pseudo-negative labels to improve pulmonary nodule detection, addressing data scarcity in medical imaging.

Findings

False positive rate reduced to 0.1266 from 0.4864

Sensitivity maintained at 0.89

Effective semi-supervised approach for medical image detection

Abstract

Generating large quantities of quality labeled data in medical imaging is very time consuming and expensive. The performance of supervised algorithms for various tasks on imaging has improved drastically over the years, however the availability of data to train these algorithms have become one of the main bottlenecks for implementation. To address this, we propose a semi-supervised learning method where pseudo-negative labels from unlabeled data are used to further refine the performance of a pulmonary nodule detection network in chest radiographs. After training with the proposed network, the false positive rate was reduced to 0.1266 from 0.4864 while maintaining sensitivity at 0.89.