Online 4-D CT Estimation for Patient-Specific Respiratory Motion Based on Real-Time Breathing Signals

Abstract

Abstract. In image-guided lung intervention, the electromagnetic (EM) tracked needle can be visualized in a pre-procedural CT by registering the EM tracking and the CT coordinate systems. However, there exist discrepancies between the static pre-procedural CT and the patient due to respiratory motion. This paper proposes an online 4-D CT estimation approach to patient-specific respiratory motion compensation. First, the motion patterns between 4-D CT data and respiratory signals such as fiducials from a number of patients are trained in a template space after image registration. These motion patterns can be used to estimate the patient-specific serial CTs from a static 3-D CT and the real-time respiratory signals of that patient, who do not generally take 4-D CTs. Specifically, the respiratory lung field motion vectors are projected onto the Kernel Principal Component Analysis (K-PCA) space, and a motion estimation model is constructed to estimate the lung field motion from the fiducial motion using the ridge regression method based on the least squares support vector machine (LS-SVM). The algorithm can be performed onsite prior to the intervention to generate the serial CT images according to the respiratory signals in advance, and the estimated CTs can be visualized in real-time during the intervention. In experiments, we evaluated the algorithm using leave-one-out strategy on 30 4-D CT data, and the results showed that the average errors of the lung field surfaces are 1.63mm.

Respiratory motion models: A review

“…This type of model can represent an average motion (Ehrhardt et al, 2008Sundaram et al, 2004) or can represent an average motion together with some information about individual variation from the average (Ehrhardt et al, 2010;Fayad et al, 2009a;He et al, 2010;Klinder et al, 2009Klinder et al, , 2010Klinder et al, , 2012Preiswerk et al, 2012). Cross-population motion models have a wide range of potential application, including quantitative analysis of motion for diagnostic purposes.…”

“…Cross-population motion models are formed from motion data acquired from many different subjects, and attempt to capture the nature of breathing motion across the population (Ehrhardt et al, , 2010(Ehrhardt et al, , 2008Fayad et al, 2009a;He et al, 2010;Klinder et al, 2009Klinder et al, , 2010Klinder et al, , 2012Preiswerk et al, 2012;Sundaram et al, 2004). This type of model can represent an average motion (Ehrhardt et al, 2008Sundaram et al, 2004) or can represent an average motion together with some information about individual variation from the average (Ehrhardt et al, 2010;Fayad et al, 2009a;He et al, 2010;Klinder et al, 2009Klinder et al, , 2010Klinder et al, , 2012Preiswerk et al, 2012).…”

Personalising population-based respiratory motion models of the heart using neighbourhood approximation based on learnt anatomical features

“…First, the estimation accuracy of the proposed personalised model was investigated by comparing it to other estimation techniques, namely no respiratory motion estimation, a standard average population-based motion model (Fayad et al, 2010;He et al, 2010;Ehrhardt et al, 2011;Klinder and Lorenz, 2012;Preiswerk et al, 2012), our previously proposed version (Peressutti et al, 2013b) and a subjectspecific motion model . The subject-specific motion model accuracy represented the best target accuracy achievable.…”

“…To this end, we compared: (I ) NOEST : no respiratory motion estimation; (II ) AVPOP : standard population-based motion model (Fayad et al, 2010;He et al, 2010;Ehrhardt et al, 2011;Klinder and Lorenz, 2012;Preiswerk et al, 2012). All N subjects were used to build an average motion model which was applied to the out-of-sample subject I oos .…”

“…To overcome the need for a subject-specific calibration scan, populationbased models have been proposed (Fayad et al, 2010;He et al, 2010;Ehrhardt et al, 2011;Klinder and Lorenz, 2012;Preiswerk et al, 2012;Samei et al, 2012). For such models, the motion in the calibration phase is estimated from calibration scans acquired previously from a sample of the population of subjects, and the model is subsequently applied to a subject not belonging to the sample.…”