Device-less gating for PET/CT using PCA

Abstract

Movement degrades image quality in PET/CT. The first step in correcting for movement is to gate the data into dif ferent motion states. The gating is usually based on information from external devices, such as the chest position for respiratory movement, or an ECG signal for cardiac gating. Various groups have proposed methods to extract a gating signal out of the PET and/or CT data. Most of these methods are slow or require prior information (and associated tuning of parameters). Here we propose and evaluate a method that uses a well-known technique for data analysis called Principal Component Analysis (PCA). We test the method on clinical PET list mode data and CINE CT images to extract a gating signal. We show good correlation with the chest position as measured by the Varian RPM system. Total processing time for PET data is less than half a minute of which most is 10 related. I. IN TRODUCTIONIn many cases in medical imaging, motion is unavoid able. For example, in diagnostic PET, acquisition duration is currently roughly 2 minutes per bed position. Respiratory motion in patients during PET acquisition leads to blurring in the resulting (static) PET images. This may in turn lead to lower detectability of tumours, inaccurate SUV calculation, and incorrect tum or planning volumes in radiation therapy [1], [2], [3], [4]. The first step towards reducing the amount of motion in the images normally involves the use of respiratory gating [5] which results in a 4-D data set, in which multiple gates containing data from the various respiratory states, are usually individually reconstructed. This is then potentially followed by motion correction.In other applications, the different motion states are of interest, for instance in cardiac studies to find the ejection fraction [6] and/or wall motion [7], or in radio-therapy [8].The gating is usually based on information from exter nal devices, such as a spyrometer or the chest position for respiratory movement, or an ECG signal for cardiac gating. Due to the extra cost and patient management associated to the additional device, but also because of some evidence of hysteresis between the internal movement and external device [9], [10], [11], various groups have proposed methods to extract a gating signal out of the PET, SPECT and/or CT data. Many groups compute the centre-of-mass (COM) in an ROI and use this as an indicator of motion, mostly of respiration, for instance in PET [12], [13] and cardiac SPECT [14]. Filtering allows separation of a respiratory and cardiac signal in cardiac PET [15] and cone-beam CINECT [16]. These methods need a high contrast region that can be tracked over time.For PET data, Visvikis et at.[17] placed an ROI over edges of boundaries (using non-attenuation corrected images) and studied the Time Activity Curve (TACs). A characteristic frequency was derived via the Fourier Transform (FT) which then allowed finding amplitude and phase images. This method worked well on phantom data with periodic movement but was not evaluated for patients. Schleyer ...

Motion Correction Strategies for Integrated PET/MR

“…The correlation with the video-based method of approximately 0.8 for both resp_pca and resp_le as reported by Thielemans et al (11,28) could be achieved only with resp_le1sens/resp_pca1sens in the present study. Possible reasons for this discrepancy include the levels of sinogram compression that were used and the sampling rates of the respiratory signals, the latter of which were 5 times higher in this study.…”

“…As PET and possibly MR data have to be gated retrospectively, exact knowledge of the corresponding respiratory state is required at any time point during the examination. Within the scope of this study, this was achieved by the acquisition of respiratory signals according to 5 methods, that is, the pressure-sensitive respiratory bellows (resp_bellows) that is shipped with the scanner by default, a prototype implementation of a self-gated MR imaging pulse sequence (resp_mr) (25,26), the PET-based sensitivity method (resp_sens) (9,10), and the application of dimensionality reduction techniques such as principal component analysis (PCA) or Laplacian eigenmaps to the sinogram space (resp_pca and resp_le, respectively) (11,27,28). Before execution of the PET-based methods, the list-mode streams were divided into nonoverlapping 200-ms time frames.…”

“…The integration of whole-body PET with MR imaging and its introduction into the clinical routine facilitates the implementation of a broad variety of motion-correction strategies, taking advantage of the respective strengths of both modalities. The respiratory signal can be obtained with external sensors, such as respiratory belts, or cameras in combination with fiducial markers mounted on the chest of the patient (1,7,8) or can be extracted from PET data (9)(10)(11). In MR, navigator echoes are used to track the position and phase shifts of objects of interest in the field of view (FOV) that are subject to respiratory motion, mostly the liver dome (12).…”

Practical PET Respiratory Motion Correction in Clinical PET/MR

“…It has been shown that a respiratory signal can be extracted from raw PET list-mode data with principal component analysis (PCA) (12). If areas of sufficient contrast in the tracer uptake in a PET scan are moving, this movement will be detectable in the raw PET data.…”

Joint PET-MR respiratory motion models for clinical PET motion correction

“…The surrogate signal for our motion model is a respiratory signal extracted using PCA from the raw PET data (Thielemans et al 2011); previously shown to provide similar information to an MR pencil-beam navigator to track breathing (Manber et al 2015). PET list-mode data are unlisted into short low spatial resolution sinogram frames.…”