Single-view 2D CNNs with Fully Automatic Non-nodule Categorization for False Positive Reduction in Pulmonary Nodule Detection
This work addresses the problem of high computational complexity and false positives in medical imaging for radiologists, though it is incremental as it builds on existing CNN techniques.
The paper tackles false positive reduction in pulmonary nodule detection by proposing an ensemble of 2D CNNs with automatic non-nodule categorization, which outperforms existing 3D CNN-based methods in computational efficiency and performance.
Background and Objective: In pulmonary nodule detection, the first stage, candidate detection, aims to detect suspicious pulmonary nodules. However, detected candidates include many false positives and thus in the following stage, false positive reduction, such false positives are reliably reduced. Note that this task is challenging due to 1) the imbalance between the numbers of nodules and non-nodules and 2) the intra-class diversity of non-nodules. Although techniques using 3D convolutional neural networks (CNNs) have shown promising performance, they suffer from high computational complexity which hinders constructing deep networks. To efficiently address these problems, we propose a novel framework using the ensemble of 2D CNNs using single views, which outperforms existing 3D CNN-based methods. Methods: Our ensemble of 2D CNNs utilizes single-view 2D patches to improve both computational and memory efficiency compared to previous techniques exploiting 3D CNNs. We first categorize non-nodules on the basis of features encoded by an autoencoder. Then, all 2D CNNs are trained by using the same nodule samples, but with different types of non-nodules. By extending the learning capability, this training scheme resolves difficulties of extracting representative features from non-nodules with large appearance variations. Note that, instead of manual categorization requiring the heavy workload of radiologists, we propose to automatically categorize non-nodules based on the autoencoder and k-means clustering.