Segmentation and tracking of lung nodules via graph‐cuts incorporating shape prior and motion from 4D CT

Abstract

Purpose: We have developed a robust tool for performing volumetric and temporal analysis of nodules from respiratory gated four-dimensional (4D) CT. The method could prove useful in IMRT of lung cancer. Methods: We modified the conventional graph-cuts method by adding an adaptive shape prior as well as motion information within a signed distance function representation to permit more accurate and automated segmentation and tracking of lung nodules in 4D CT data. Active shape models (ASM) with signed distance function were used to capture the shape prior information, preventing unwanted surrounding tissues from becoming part of the segmented object. The optical flow method was used to estimate the local motion and to extend three-dimensional (3D) segmentation to 4D by warping a prior shape model through time. The algorithm has been applied to segmentation of wellcircumscribed, vascularized, and juxtapleural lung nodules from respiratory gated CT data. Results: In all cases, 4D segmentation and tracking for five phases of high-resolution CT data took approximately 10 min on a PC workstation with AMD Phenom II and 32 GB of memory. The method was trained based on 500 breath-held 3D CT data from the LIDC data base 1 and was tested on 17 4D lung nodule CT datasets consisting of 85 volumetric frames. The validation tests resulted in an average Dice Similarity Coefficient (DSC) = 0.68 for all test data. An important by-product of the method is quantitative volume measurement from 4D CT from end-inspiration to end-expiration which will also have important diagnostic value. Conclusion: The algorithm performs robust segmentation of lung nodules from 4D CT data. Signed distance ASM provides the shape prior information which based on the iterative graph-cuts framework is adaptively refined to best fit the input data, preventing unwanted surrounding tissue from merging with the segmented object.

3D Hermite Transform Optical Flow Estimation in Left Ventricle CT Sequences

“…Yoon et al [37] presented a method for motion estimation applied to cone-beam CT, their work uses an energy functional, which includes as terms: a data fidelity, a regularization term, and the optical flow restriction. On the other hand, Jungwon et al [38] used the optical flow estimation to calculate the local motion, allowing a 3D segmentation extension. Their model includes a shape distortion over time term, allowing segmenting and tracking the lung nodules.…”

Segmentation and classification of lung nodules from Thoracic CT scans : methods based on dictionary learning and deep convolutional neural networks.

“…The reconstructed shape, shown in bottom right of Fig. 6, is not sparse in this case -this reconstruction required more than 80% of the dictionary atoms in the linear approximation for the same value of λ in (9). Despite the large number of atoms used, the input shape cannot be accurately captured by the dictionary.…”

“…The authors segmented neuronal structures in electron microscopic and cell segmentation in light microscopic images. The 2-D operators in [9] were replaced with their 3-D counterparts in [16] and breast cancer metastasis detection for sentinel lymph nodes. In [51], diagnosis of Alzheimer disease was performed based on MRI and PET modalities.…”

Segmentation, tracking, and kinematics of lung parenchyma and lung tumors from 4D CT with application to radiation treatment planning.

“…Results We performed segmentation of nine 4-D patient data with well-circumscribed or vascularized lung nodules [30,31]. Figure 22 shows a 2-D axial slice and a 2-D coronal slice for phase 10 of a patient from this category.…”

“…Testing step After obtaining the warped mean shape, we perform 4-D (3-D+t) segmentation via graph cuts through five phases with shape refinement via PCA analysis [30,31]. The training step was discussed in Section 3.6.2.…”