Scatter fraction: measurement and correction

The concept of scatter in Positron Emission Tomography is reviewed regarding origin and influence on data. Different ways to measure and correct for scatter are discussed.This article was presented at the 1st EEC workshop on accuracy determination in PET, January 19–20th. 1989 Pisa, Italy (COMACBME Concerted Project Characterization and Standardization of PET Instrumentation)     

Scatter modelling and compensation in emission tomography

In nuclear medicine, clinical assessment and diagnosis are generally based on qualitative assessment of the distribution pattern of radiotracers used. In addition, emission tomography (SPECT and PET) imaging methods offer the possibility of quantitative assessment of tracer concentration in vivo to quantify relevant parameters in clinical and research settings, provided accurate correction for the physical degrading factors (e.g. attenuation, scatter, partial volume effects) hampering their quantitative accuracy are applied. This review addresses the problem of Compton scattering as the dominant photon interaction phenomenon in emission tomography and discusses its impact on both the quality of reconstructed clinical images and the accuracy of quantitative analysis. After a general introduction, there is a section in which scatter modelling in uniform and non-uniform media is described in detail. This is followed by an overview of scatter compensation techniques and evaluation strategies used for the assessment of these correction methods. In the process, emphasis is placed on the clinical impact of image degradation due to Compton scattering. This, in turn, stresses the need for implementation of more accurate algorithms in software supplied by scanner manufacturers, although the choice of a general-purpose algorithm or algorithms may be difficult.     

Development of a practical image-based scatter correction method for brain perfusion SPECT: comparison with the TEW method

Purpose An image-based scatter correction (IBSC) method was developed to convert scatter-uncorrected into scatter-corrected SPECT images. The purpose of this study was to validate this method by means of phantom simulations and human studies with ^99mTc-labeled tracers, based on comparison with the conventional triple energy window (TEW) method.Methods The IBSC method corrects scatter on the reconstructed image with Changs attenuation correction factor. The scatter component image is estimated by convolving with a scatter function followed by multiplication with an image-based scatter fraction function. The IBSC method was evaluated with Monte Carlo simulations and ^99mTc-ethyl cysteinate dimer SPECT human brain perfusion studies obtained from five volunteers. The image counts and contrast of the scatter-corrected images obtained by the IBSC and TEW methods were compared.Results Using data obtained from the simulations, the image counts and contrast of the scatter-corrected images obtained by the IBSC and TEW methods were found to be nearly identical for both gray and white matter. In human brain images, no significant differences in image contrast were observed between the IBSC and TEW methods.Conclusion The IBSC method is a simple scatter correction technique feasible for use in clinical routine.     

Transmission-based scatter correction of 180° myocardial single-photon emission tomographic studies

Meaningful comparison of single-photon emission tomographic (SPET) reconstructions for data acquired over 180° or 360° can only be performed if both attenuation and scatter correction are applied. Convolution subtraction has appeal as a practical method for scatter correction; however, it is limited to data acquired over 360°. A new algorithm is proposed which can be applied equally well to data acquired over 180° or 360°. The method involves estimating scatter based on knowl edge of reconstructed transmission data in combination with a reconstructed estimate of the activity distribution, obtained using attenuation correction with broad beam attenuation coefficients. Processing is implemented for planes of activity parallel to the projection images for which a simplified model for the scatter distribution may be applied, based on the measured attenuation. The appropriate broad beam (effective) attenuation coefficients were determined by considering the scatter buildup equation. It was demonstrated that narrow beam attenuation coefficients should be scaled by 0.75 and 0.65 to provide broad beam attenuation coefficients for technetium-99m and thallium-201 respectively. Using a thorax phantom, quantitative accuracy of the new algorithm was compared with conventional transmission-based convolution subtraction (TDCS) for 360° data. Similar heart to lung contrasts were achieved and correction of 180° data yielded a 10.4% error for cardiac activity compared to 5.2% for TDCS. Contrast for myocardium to ventricular cavity was similarly good for scatter-corrected 180° and 360° data, in contrast to attenuation-corrected data, where contrast was significantly reduced. The new algorithm provides a practical method for correction of scatter applicable to 180° myocardial SPET.     

A method for calibrating three-dimensional positron emission tomography without scatter correction

Calibration for three-dimensional positron emission tomography (3D PET) using a uniform cylinder and cross-calibration with aliquots requires correction for scatter and attenuation. Thus the accuracy of thecalibration is dependent on the scatter correction method, and on the applicability of the scatter correction for different regions of the body. A method has been developed which provides a calibration which does not require correction for scatter or attenuation, making it generally applicable and independent of the scatter correction. The method has been previously described for measurement of the absolute sensitivity of tomography devices. This approach has been extended to give a calibration of the PET camera in air in units of kBq/pixel. The reconstructed images are multiplied by this factor to, give accurate activity concentrations, after attenuation and scatter correction. The method has been used with a fully 3D filtered back-projection (reprojection) algorithm and iterative convolution-subtraction scatter correction on data from an ECAT 953B. Using this method 3D PET images have been calibrated te, within ±5% accuracy, but this is highly dependent on the accuracy of the scatter correction. The method described here is practical and provides a means of calibrating a 3D PET system without the need for correction for scatter or attenuation of the calibration data.     

Validation of quantitative brain dopamine D2 receptor imaging with a conventional single-head SPET camera

Phantom measurements were performed with a conventional single-head single-photon emission tomography (SPET) camera in order to validate the relevance of the basal ganglia/frontal cortex iodine-123 iodobenzamide (IBZM) uptake ratios measured in patients. Inside a cylindrical phantom (diameter 22 cm), two cylinders with a diameter of 3.3 cm were inserted. The activity concentrations of the cylinders ranged from 6.0 to 22.6 kBq/ml and the cylinder/background activity ratios varied from 1.4 to 3.8. From reconstructed SPET images the cylinder/ background activity ratios were calculated using three different regions of interest (ROIs). A linear relationship between the measured activity ratio and the true activity ratio was obtained. In patient studies, basal ganglia/frontal cortex IBZM uptake ratios determined from the reconstructed slices using attenuation correction prior to reconstruction were 1.30±0.03 in idiopathic Parkinson's disease (n=9), 1.33±0.09 in infantile and juvenile neuronal ceroid lipofuscinosis (n=7) and 1.34±0.05 in narcolepsy (n=8). Patients with Huntington's disease had significantly lower ratios (1.09±0.04, n=5). The corrected basal ganglia/frontal cortex ratios, determined using linear regression, were about 80% higher. The use of dual-window scatter correction increased the measured ratios by about 10%. Although comprehensive correction methods can further improve the resolution in SPET images, the resolution of the SPET system used by us (1.52 cm) will determine what is achievable in basal ganglia D2 receptor imaging.Paper presented in part at the European Association of Nuclear Medicine Congress, 22–26 August 1992, Lisbon, Portugal Correspondence to: P. Nikkinen, Division of Nuclear Medicine, Meilahti Hospital, SF-00290 Helsinki, Finland     

Efficient scatter modelling for incorporation in maximum likelihood reconstruction

Definition of a simplified model of scatter which can be incorporated in maximum likelihood reconstruction for single-photon emission tomography (SPET) continues to be appealing; however, implementation must be efficient for it to be clinically applicable. In this paper an efficient algorithm for scatter estimation is described in which the spatial scatter distribution is implemented as a spatially invariant convolution for points of constant depth in tissue. The scatter estimate is weighted by a space-dependent build-up factor based on the measured attenuation in tissue. Monte Carlo simulation of a realistic thorax phantom was used to validate this approach. Further efficiency was introduced by estimating scatter once after a small number of iterations using the ordered subsets expectation maximisation (OSEM) reconstruction algorithm. The scatter estimate was incorporated as a constant term in subsequent iterations rather than modifying the scatter estimate each iteration. Monte Carlo simulation was used to demonstrate that the scatter estimate does not change significantly provided at least two iterations OSEM reconstruction, subset size 8, is used. Complete scatter-corrected reconstruction of 64 projections of 40×128 pixels was achieved in 38 min using a Sun Sparc20 computer. Received 11 August and in revised form 24 August 1998     

Assessment of skull base involvement in nasopharyngeal carcinoma: comparisons of single-photon emission tomography with planar bone scintigraphy and X-ray computed tomography

The diagnostic contribution of single-photon emission tomography (SPET) to the detection of bone lesions of the skull base was explored in 200 patients with nasopharyngeal carcinoma (NPC). Comparison of SPET with planar bone scintigraphy showed that SPET improved the contrast and better defined the lesions in 107 out of the 200 patients. Comparison of SPET with X-ray computed tomography (CT) showed that SPET did not miss the lesions detected by CT while CT missed 49% of the lesions detected by SPET. The only false-positive lesion with SPET was detected in the mastoid bone. SPET detected skull base lesions in all of the 35 patients with cranial nerve involvement, while CT missed eight and planar bone scintigraphy missed four. The findings suggest that SPET should be included in the routine check-up examinations of patients with NPC.     

External reference markers for the correction of head rotation in brain single-photon emission tomography

Accurate reorientation of brain single-photon emission tomography (SPET) is required for quantitative procedures and for correlation with other imaging modalities. Traditionally, brain SPET has utilized reoriented slices parallel to the orbitomeatal line (OML). Reorientation using internal landmarks would be more convenient but has not been systematically compared with the use of external landmarks. We compared the interobserver reproducibility for defining the sagittal and coronal angular deviations using internal landmarks, a visual method based upon external reference markers, and an automated method based upon external reference markers. Internal landmarks were inaccurate for defining the OML whether this was based upon the frontal-occipital or frontal-cerebellar plane. External reference markers resulted in significantly lower interobserver differences for both sagittal and coronal reorientation. An operator-independent implementation proved to be feasible and provided an objective measure of marker coplanarity. In summary, external reference markers should be used when reproducible reorientation and ROI placement are required.     

Efficient object scatter correction algorithm for third and fourth generation CT scanners

X-ray photons which are scattered inside the object slice and reach the detector array increase the detected signal and produce image artifacts as “cupping” effects in large objects and dark bands between regions of high attenuation. The artifact amplitudes increase with scanned volume or slice width. Object scatter can be reduced in third generation computed tomography (CT) geometry by collimating the detector elements. However, a correction can still improve image quality. For fourth generation CT geometry, only poor anti-scatter collimation is possible and a numeric correction is necessary. This paper presents a correction algorithm which can be parameterized for third and fourth generation CT geometry. The method requires low computational effort and allows flexible application to different body regions by simple parameter adjustments. The object scatter intensity which is subtracted from the measured signal is calculated with convolution of the weighted and windowed projection data with a spatially invariant “scatter convolution function“. The scatter convolution function is approximated for the desired scanner geometry from pencil beam simulations and measurements using coherent and incoherent differential scatter cross section data. Several examples of phantom and medical objects scanned with third and fourth generation CT systems are discussed. In third generation scanners, scatter artifacts are effectively corrected. For fourth generation geometry with poor anti-scatter collimation, object scatter artifacts are strongly reduced. Received: 13 March 1997; Revision received: 1 December 1997; Accepted: 11 May 1998     

Verification of the integral transformation of the projections technique for scatter correction in positron tomographs

A scatter correction algorithm, based on the integral transformation of the projections, has been implemented when using the C.N.R. positron tomograph in Pisa. The performance of the method has been evaluated by measurements on ad hoc phantoms. The technique allows a significant average reduction of the reconstructed scatter fraction. The limitations of this approach have also been investigated. Correspondence to: P. Prati     

Attenuation correction improves the detection of viable myocardium by thallium-201 cardiac tomography in patients with previous myocardial infarction and left ventricular dysfunction

The aim of this study was to determine the influence of attenuation-corrected thallium-201 stress/redistribution/reinjection single-photon emission tomography (SPET) on the number of viable segments in patients with previous myocardial infarction and dysfunctional myocardium. Fifty-one patients with previous myocardial infarction and left ventricular dysfunction were included in the study. In all patients, ²⁰¹Tl non-corrected (NC) and attenuation-corrected (AC) SPET was performed using a stress/redistribution/reinjection protocol followed by coronary angiography. A semiquantitative analysis was performed using polar maps for NC and AC stress, redistribution and reinjection short-axis and vertical long-axis (apex) slices. Severe (perfusion defect below 50%/maximal count rate: PD_<50), mild and moderate persistent defects for redistribution and reinjection were evaluated for both NC and AC studies. A total of 1581 segments were evaluated by semiquantitative segmental analysis for both NC and AC studies for each redistribution and reinjection map. In the redistribution maps, NC revealed a total of 352 segments and AC a total of 222 segments with impaired perfusion below 50% of the maximal count rate (PD_<50). The mean number of affected segments was 6.9±5.5 in the case of NC and 4.4±4.8 in the case of AC (P<0.001). In the reinjection maps, NC revealed a total of 263 non-viable segments (PD_<50) and AC a total of 169 non-viable segments. The mean number of affected segments was 5.2±5.3 in the case of NC and 3.3±4.2 in the case of AC (P<0.001). Recovery of function was better predicted by AC than by NC in 20% of patients in the follow-up group. Therefore, the use of attenuation correction influences the extent of viable segments by showing more viable segments in either redistribution or reinjection maps. ²⁰¹Tl imaging without attenuation correction may underestimate the extent of tissue viability, which may contribute to the lower sensitivity compared to fluorine-18-fluorodeoxyglucose positron emission tomography, where attenuation correction is a routinely performed procedure. Received 26 October and in revised form 23 December 1998     

A new thresholding method for volume determination by SPECT

The quantification of organ volumes from SPECT images suffers from two major problems: image segmentation and imperfect system transfer function. Image segmentation defines the borders of an organ and allows volume measurements by counting the voxels inside this contour in all slices containing parts of this organ. A review of the literature, showed that several investigators use a fixed threshold (FT) to determine the organ pixels. It is our aim to demonstrate that the threshold has to be adapted to every single case because its value is dependent upon several factors, such as size and contrast. Therefore a threshold selection algorithm, based on the gray level histogram (GLH), is evaluated. It is nearly impossible to calculate and eliminate errors induced by the complex system response function. A correction method based on linear regression is proposed. By minimizing the relative error (), a linear correlation (Y=AX+B) between the true volume (Y) and the measured volume (X) is established for three fixed thresholds (30%, 40%, 50%) and for the GLH method. The methods are evaluated on a series of nineteen phantoms with a volume range between 9.8 and 202.5 ml. The relative error is minimal for the GLH method. The whole procedure is semi-automated and virtually operator independent.     

Transmission scanning in emission tomography

Attenuation correction in single-photon (SPET) and positron emission (PET) tomography is now accepted as a vital component for the production of artefact-free, quantitative data. The most accurate attenuation correction methods are based on measured transmission scans acquired before, during, or after the emission scan. Alternative methods use segmented images, assumed attenuation coefficients or consistency criteria to compensate for photon attenuation in reconstructed images. This review examines the methods of acquiring transmission scans in both SPET and PET and the manner in which these data are used. While attenuation correction gives an exact correction in PET, as opposed to an approximate one in SPET, the magnitude of the correction factors required in PET is far greater than in SPET. Transmission scans also have a number of other potential applications in emission tomography apart from attenuation correction, such as scatter correction, inter-study spatial co-registration and alignment, and motion detection and correction. The ability to acquire high-quality transmission data in a practical clinical protocol is now an essential part of the practice of nuclear medicine. Received: 19 February 1998 / Accepted: 19 March 1998     

Simultaneous dual-isotope technetium-99m/thallium-201 cardiac SPET imaging using a projection-dependent spilldown correction factor

A spilldown correction method is proposed for the thallium-201 window image in simultaneous dual-isotope technetium-99m/thallium-201 single-photon emission tomographic (SPET) imaging based on a single acquisition into three energy windows. In this method, images are simultaneously acquired in two standard energy windows over the^99mTc and²⁰¹Tl photopeak regions and a third spilldown window adjacent to the²⁰¹Tl window. Using a Monte Carlo simulation of SPET, the fractional amount of^99mTc and²⁰¹Tl spilldown in the²⁰¹Tl window with respect to the total counts from the spilldown window, k12, was calculated for simulated images of point sources at varying depths within a water-filled elliptical tub phantom. When applied to experimental acquisitions, k12, multiplied by the total counts from the spilldown window, is then subtracted from the²⁰¹Tl window image to produce the corrected image. However, for successful applications in SPET, k12 must be determined on a projection-by-projection basis since k12 is depth dependent. Thus, a regression relation was obtained between k12 and the total count ratio of the spilldown to^99mTc windows, k23. The spilldown correction method was applied to²⁰¹T1 photopeak images of an extended source distribution in uniform and nonuniform attenuating media with dual-isotope^99mTc/²⁰¹Tl and single-isotope²⁰¹Tl. A marked improvement in image contrast was observed between the corrected and uncorrected²⁰¹Tl window images. The average count ratio of uncorrected dual-isotope²⁰¹Tl/single-isotope²⁰¹Tl was 3.08 for uniform and 2.99 for non-uniform attenuating media. When corrected, this value approached unity (0.97 and 1.08) and the corrected dual-isotope images were visually similar The presented method in which spilldown correction for the²⁰¹Tl photopeak window is performed in a single acquisition on a projection-by-projection basis in simultaneous dual-isotope SPET imaging could form the basis for a rapid and accurate clinical dual-isotope protocol independent of radiotracer administration.     

Quantitation in planar renal scintigraphy: which µ value should be used?

The attenuation coefficient value μ used by different authors for quantitation in planar renal scintigraphy varies greatly, from the theoretical value of 0.153 cm^–1 (appropriate for scatter-free data) down to 0.099 cm^–1 (empirical value assumed to compensate for both scatter and attenuation). For a 6-cm-deep kidney, such variations introduce up to 30% differences in absolute measurement of kidney activity. Using technetium-99m phantom studies, we determined the μ values that would yield accurate kidney activity quantitation for different energy windows corresponding to different amounts of scatter, and when using different image analysis approaches similar to those used in renal quantitation. With the 20% energy window, it was found that the μ value was strongly dependent on the size of the region of interest (ROI) and on whether background subtraction was performed: the μ value thus varied from 0.119 cm^–1 (loose ROI, no background subtraction) to 0.150 cm^–1 (kidney ROI and background subtraction). When using data from an energy window that could be considered scatter-free, the μ value became almost independent of the image analysis scheme. It is concluded that: (1) when performing background subtraction, which implicitly reduces the effect of scatter, the μ value to be used for accurate quantitation is close to the theoretical μ value; (2) if the acquired data were initially corrected for scatter, the appropriate μ value would then be the theoretical μ value, whatever the image analysis scheme. Received 15 July and in revised form 25 August 1999     

Quantitative analysis in single photon emission tomography (SPET)

Quantitative analysis can improve the sensitivity and specificity of single photon emission tomography (SPET) procedures, as well as reduce inter- and intraobserver variabilities. Quantification of the radioactivity distribution is the ultimate goal of SPCT. In this review we consider the basic requirements for an optimum three-dimensional reconstruction of the radionuclide distribution to enable quantification. Attenuation and scatter correction as well as varying resolution are the major problems. In the older SPET systems quantification was hampered by the lack of system sensitivity and sufficient computer power. Therefore, the imaging system was often assumed to be shift invariant and linear and the attenuation throughout the object uniform. More sophisticated solutions have been proposed and with more or less success implemented, but not for application in daily practice. Knowledge (measurement) of the attenuation is often required. New generation SPET systems employing multi-detectors and super minicomputers will ease the implementation of these solutions. Offprint requests to: J.A.K. Blokland     

Resting state rCBF mapping with single-photon emission tomography and positron emission tomography: magnitude and origin of differences

Single-photon emission tomography (SPET), using technetium-99m hexamethylpropylene amine oxime, and positron emission tomography (PET), using oxygen-15 butanol were compared in six healthy male volunteers with regard to the mapping of resting state regional cerebral blood flow (rCBF). A computerized brain atlas was utilized for 3D regional analyses and comparison of 64 selected and normalized volumes of interest (VOIs). The normalized mean rCBF values in SPET, as compared to PET, were higher in most of the Brodmann areas in the frontal and parietal lobes (4.8% and 8.7% respectively). The average differences were small in the temporal (2.3%) and occipital (1.1%) lobes. PET values were clearly higher in small VOIs like the thalamus (12.3%), hippocampus (12.3%) and basal ganglia (9.9%). A resolution phantom study showed that the in-plane SPET/PET system resolution was 11.0/7.5 mm. In conclusion, SPET and PET data demonstrated a fairly good agreement despite the superior spatial resolution of PET. The differences between SPET and PET rCBF are mainly due to physiological and physical factors, the data processing, normalization and co-registration methods. In order to further improve mapping of rCBF with SPET it is imperative not only to improve the spatial resolution but also to apply accurate correction techniques for scatter, attenuation and non-linear extraction. Received 3 August and in revised form 1 October 1997     

Lateral collateral ligament tear of the knee: appearances on bone scintigraphy with single-photon emission tomography

We report two cases, with magnetic resonance imaging correlation, of acute lateral collateral ligament tear of the knee following trauma with findings on bone scintigraphy with single-photon emission tomography (SPET). The typical bone scan features are presented. In addition, the advantages of the use of SPET in the detection of associated lesions are discussed.