Surface deformation analysis of collapsed lungs using model-based shape matching

Abstract

PurposeTo facilitate intraoperative localization of lung nodules, this study used model-based shape matching techniques to analyze the inter-subject three-dimensional surface deformation induced by pneumothorax. Methods: Contrast- enhanced computed tomography (CT) images of the left lungs of 11 live beagle dogs were acquired at two bronchial pressures (14 and 2 cm). To address shape matching problems for largely deformed lung images with pixel intensity shift, a complete Laplacian-based shape matching solution that optimizes the differential displacement field was introduced.ResultsExperiments were performed to confirm the methods’ registration accuracy using CT images of lungs. Shape similarity and target displacement errors in the registered models were improved compared with those from existing shape matching methods. Spatial displacement of the whole lung’s surface was visualized with an average error of within 5 mm.ConclusionThe proposed methods address problems with the matching of surfaces with large curvatures and deformations and achieved smaller registration errors than existing shape matching methods, even at the tip and ridge regions. The findings and inter-subject statistical representation are directly available for further research on pneumothorax deformation modeling.Electronic supplementary materialThe online version of this article (10.1007/s11548-019-02013-0) contains supplementary material, which is available to authorized users.

Model-based registration for pneumothorax deformation analysis using intraoperative cone-beam CT images

“…STEP 1 Affine Transformation STEP 2 Piecewise Affine Transformation [11] STEP 3 Laplacian-based Shape Matching [7] First, in STEP 1, the rotation and scale are adjusted, and then, in STEP 2, Piecewise Affine Transform (PWA), which performs Affine Transformation on each of the divided regions, is employed. Here, the source is updated to minimize the objective function defined by Eq.…”

“…where d is the mean distance [7], mean value of the nearest bidirectional point-to-surface distance of the two meshes S D and S I . w is a weight, and p D and p I are the position of surgical clips placed on the lung surface in the inflated and deflated states, respectively.…”

“…The discrete Laplacian is a shape descriptor, and by using it, minimizes shape differences while maintaining shape. Regarding the details of the algorithms and the performance, see [7].…”

“…Examples of studies that performed registration for organ deformation include the investigations that performed preoperative CT and intraoperative CT registration for the liver and lung [3][4] and a study that performed registration during the respiratory cycle [5]. 3D analyses of pneumothorax deformation of animal lungs have been conducted [6] [7].…”

Multimodal Patient-Specific Registration for Breast Imaging Using Biomechanical Modeling with Reference to AI Evaluation of Breast Tumor Change

“…By assigning reasonable tissue property and boundary conditions, finite element analysis can provide realistic deformation predictions. Biomechanical models have been used to simulate the deformation of the lung [ 11 ] and the breast such as compression in X-ray mammography [ 12 , 13 ], and behavior under gravity from prone to supine positions [ 5 , 14 , 15 ]. Han et al developed a homogeneous model to predict large breast deformations from the prone position to supine position [ 15 ], but the mechanical property variance has not been considered.…”