Pathological Pulmonary Lobe Segmentation from CT Images using Progressive Holistically Nested Neural Networks and Random Walker
This work addresses the clinically important problem of enabling regional lung disease analysis for medical professionals, representing a novel application of deep learning to this task.
The authors tackled the problem of automatic pathological pulmonary lobe segmentation from CT images, which is challenging due to incomplete boundaries, by proposing a novel method that combines progressive holistically-nested neural networks with the random walker algorithm, achieving a high mean Jaccard score of 0.888±0.164 on a held-out set of 154 CT scans and significantly outperforming a state-of-the-art method.
Automatic pathological pulmonary lobe segmentation(PPLS) enables regional analyses of lung disease, a clinically important capability. Due to often incomplete lobe boundaries, PPLS is difficult even for experts, and most prior art requires inference from contextual information. To address this, we propose a novel PPLS method that couples deep learning with the random walker (RW) algorithm. We first employ the recent progressive holistically-nested network (P-HNN) model to identify potential lobar boundaries, then generate final segmentations using a RW that is seeded and weighted by the P-HNN output. We are the first to apply deep learning to PPLS. The advantages are independence from prior airway/vessel segmentations, increased robustness in diseased lungs, and methodological simplicity that does not sacrifice accuracy. Our method posts a high mean Jaccard score of 0.888$\pm$0.164 on a held-out set of 154 CT scans from lung-disease patients, while also significantly (p < 0.001) outperforming a state-of-the-art method.