Automatic Segmentation of Pulmonary Lobes Using a Progressive Dense V-Network

Abstract

Reliable and automatic segmentation of lung lobes is important for diagnosis, assessment, and quantification of pulmonary diseases. The existing techniques are prohibitively slow, undesirably rely on prior (airway/vessel) segmentation, and/or require user interactions for optimal results. This work presents a reliable, fast, and fully automated lung lobe segmentation based on a progressive dense V-network (PDV-Net). The proposed method can segment lung lobes in one forward pass of the network, with an average runtime of 2 seconds using 1 Nvidia Titan XP GPU, eliminating the need for any prior atlases, lung segmentation or any subsequent user intervention. We evaluated our model using 84 chest CT scans from the LIDC and 154 pathological cases from the LTRC datasets. Our model achieved a Dice score of 0.939 ± 0.02 for the LIDC test set and 0.950 ± 0.01 for the LTRC test set, significantly outperforming a 2D U-net model and a 3D dense V-net. We further evaluated our model against 55 cases from the LOLA11 challenge, obtaining an average Dice score of 0.935-a performance level competitive to the best performing team with an average score of 0.938. Our extensive robustness analyses also demonstrate that our model can reliably segment both healthy and pathological lung lobes in CT scans from different vendors, and that our model is robust against configurations of CT scan reconstruction.

Lung Lobe Segmentation Based on Lung Fissure Surface Classification Using a Point Cloud Region Growing Approach

“…To further verify the performance of the proposed method, we compared the average score from each lung lobe between our method and two representative AI-based methods (a sequence of convolutional neural networks for marginal learning [20] and progressive dense V-Network [4]). The quantitative evaluation results on each lung lobe are shown in Table 2 (LobeNet_V2 is the name of the net in [20] given by authors for LOLA11).…”

“…George et al used progressive holistically nested neural networks to predict lung lobe boundaries in 2D slices, and then combined them with the 3D random walker to achieve lung lobe segmentation [3]. Imran et al proposed a fast and reliable progressive dense V-network [4]. In order to prevent over fitting, Ferreira et al integrated several regularization methods on the basis of a residual network [5].…”

Surrogate Supervision for Medical Image Analysis: Effective Deep Learning From Limited Quantities of Labeled Data

“…Lung lobe segmentation requires assigning every voxel in a chest CT to one of the 5 major lobes of the lung or background. For this, we use the progressive dense V-Net introduced in [14]. Since thorax geometry is consistent in chest CT scans (assuming images are re-oriented to a common axis code), it makes sense to use rotation as surrogate supervision.…”

Semi-supervised Multi-task Learning with Chest X-Ray Images

“…Following [15], with deep-supervision, APPAU-Net generates side-outputs at different resolutions from the decoder. The side-outputs are progressively added to the next side-outputs before reaching the final segmentation at the original image resolution.…”