Eric G. Hawman

A method for improving the efficiency of myocardial perfusion imaging using conventional SPECT and SPECT/CT imaging systems

Siemens has developed a new IQ•SPECT™ product to improve the efficiency of myocardial perfusion imaging (MPI) using conventional large-field-of-view SPECT and SPECT/CT systems. In this article we present the key technology components that enable this product to perform MPI in less than 5 minutes or at an equivalently lower dose. The enabling hardware is a specially designed variable-focus collimator for cardiac imaging. Images are acquired with the collimator mounted on a Symbia SPECT/CT system rotating about the patient in a cardio-centric orbit at a fixed radius of 28 cm. The acquired data are reconstructed using an iterative reconstruction technique employing the conjugate-gradient method with the Mighell chi-square objective function accounting for Poisson statistics. Each collimator is characterized by measuring the orientations of its holes to account for the deviations from design specifications that are introduced in the casting process. In addition, the 3D point-response function (PRF) is modeled from the autocorrelation of the hexagonal shapes of the collimator holes at the entrance and exit sides. This PRF is no longer just an approximate Gaussian but is more conical at the distances of interest. The system matrix accounts for the deflection of the heads as they rotate about the patient. The deflections were measured for a number of systems using an Optotrak™ optical fixture to obtain accurate 3D orbit information for each head. The reconstruction engine applies the flood-field uniformity corrections (instead of being applied to the raw data) and also estimates patient motion vectors from the distortion-corrected projection images. Attenuation compensation is applied using a patient-specific CT-derived mu map, and an energy-window-based estimate is used to correct for patient-induced scatter. Phantom and patient studies are presented to demonstrate the diagnostic quality of the images acquired using fast or low-dose protocols.

Image Noise, Resolution, and Lesion Detectability in Single Photon Emission CT

Single Photon Emission Computed Tomography (SPECT) based on rotational scintillation gamma cameras in combination with filtered back-projection allows full 3-D imaging of isotope uptake in an organ volume. (1, 2,3,4) This paper treats the physical dependence of lesion detectability in SPECT imaging on the lesion uptake, lesion size, object size, and the number of photons collected. The lesion contrast depends on both camera resolution and reconstruction filtering. The relative photon-limited image noise is proportional to the three halves power of the number of linear samples covering the object, inversely proportional to the square root of the number of photons collected, and is also affected by the reconstruction filter chosen. Based on signal-to-noise analysis, the total number of counts necessary for lesion detection is determined.

Digital Boundary Detection Techniques For The Analysis Of Gated Cardiac Scintigrams

Boundary detection in conventional nuclear medicine scintigrams is often difficult for several reasons. First, scintigrams generally have a low signal-to-noise ratio. Second, edge structures are poorly defined because of the low resolution of gamma ray cameras; and finally, edge contrast is usually reduced by foreground and background activity. In this paper we report on heuristic approaches we have taken to solve these problems and to develop programs for the display of cardiac wall motion and for the automatic determination of left ventricular ejection fraction. Our approach to processing cardiac scintigrams entails several steps: smoothing, edge enhancement, and contour extraction. We discuss each of these steps in light of the goal of producing cardiac boundaries which are spatially and temporally smooth and continuous. Boundary detection results are presented for some selected clinical images.

Statistically based spatially adaptive noise reduction of planar nuclear studies

The data-driven Pixon noise-reduction method is applied to nuclear studies. By using the local information content, it preserves all statistically justifiable image features without generating artifacts. Statistical measures provide the user a feedback to judge if the processing parameters are optimal. The present work focuses on planar nuclear images with known Poisson noise characteristics. Its ultimate goals are to: (a) increase sensitivity for detection of lesions of small size and/or of small activity-to-background ratio, (b) reduce data acquisition time, and (c) reduce patient dose. Data are acquired using Data Spectrum’s cylinder phantom in two configurations: (a) with hot and cold rod inserts at varying total counts and (b) with hot sphere inserts at varying activity-to-background ratios. We show that the method adapts automatically to both hot and cold lesions, concentration ratios, and different noise levels and structure dimensions. In clinical applications, slight adjustment of the parameters may be needed to adapt to the specific clinical protocols and physician preference. Visually, the processed images are comparable to raw images with ~16 times as many counts, and quantitatively the reduced noise equals that obtained with ~50 times as many counts. We also show that the Pixon method allows for identification of spheres at low concentration ratios, where raw planar imaging fails and matched filtering underperforms. Conclusion: The Pixon method significantly improves the image quality of data at either reduced count levels, or low target-to-background ratios. An analysis of clinical studies is now warranted to assess the clinical impact of the method.

A Method of Improving Photon Detection Efficiency of High Purity Germanium Cameras by including Double-Interaction Photopeak Events

We propose a method to utilize double-interaction photopeak events in high purity Germanium cameras to improve sensitivity with no apparent degradation in spatial resolution. This procedure is based on an energy comparison between the energy deposited at two interaction sites for photons of up to 255 keV. To determine the sensitivity improvement quantitatively, we simulated the photon transport process for 140 keV photons in detail for Ge detectors having thicknesses of 1.00 to 2.00 cm in 0.25 cm steps. Discrete HPGe cameras were modeled using pixel sizes of (1.0 mm)2, (1.5 mm)2, and (2.0 mm)2. Our results indicate sensitivity improvements of 29-42%, depending on detector thickness and pixel size.

An Attenuation Correction System for a Dedicated Small FOV, Dual Head, Fixed-90°, Cardiac Gamma Camera Using Arrays of Gd-153 Line Sources

A new transmission source-based attenuation correction system, c.clear, has been developed for the Siemens c.cam cardiac SPECT camera. The transmission sources are two arrays of 14 Gd-153 line sources placed in cassette wings affixed to the detectors. The source strengths decrease geometrically away from the array center. The absorbed radiation dose received from these sources is about 0.01% of that received from a conventional Tl-201 myocardial scan. The line sources are stationary relative to the detector during the SPECT scan. The emission and transmission radiation are acquired simultaneously and are inherently registered spatially and temporally, so avoiding misregistration artifacts. Transmission reconstruction is performed using an ordered subset maximum likelihood algorithm with gradient descent. It models the geometry of the sources and the radiation emanating from these sources as narrow fan beams. The parameters of this model are estimated from near and far blank calibration images. It includes correction for the Compton emission down-scatter into the transmission energy window that occurs during the simultaneous Tc-99m and Gd-153 imaging. The down-scatter is estimated from down-scatter measured for a short interval when the line sources are obstructed by shutters in the cassette wings. The down-scatter is estimated by linear filtering using the measured short-interval down-scatter and the emission data. The transverse FOV of the camera can truncate the transmission radiation, particularly for obese or improperly positioned patients. Compensation for such truncations is achieved by employing robust data extrapolation techniques and extending the model of the source geometry beyond the measured FOV. This attenuation correction method has been investigated using both phantom and clinical studies and has been shown to produce corrected emission reconstructions of excellent quality.

Adaptive noise reduction and sharpening of OSEM-reconstructed data

The Pixon method, a statistically rigorous procedure for adaptive noise suppression that avoids the generation of spurious artifacts yet preserves all the statistically justifiable image features resident in the raw counts, is applied to nuclear studies. The present work focuses on adaptive postsmoothing and sharpening of OSEM-reconstructed data at various count levels, with the ultimate goals to (i) increase sensitivity for detection of lesions of small size and/or of small activity-to-background ratio, (ii) reduce data acquisition time, and (iii) reduce patient dose. We use simulated and measured data and human-observer studies, which are analyzed using quantitative measures. The detectability shows improvement, as does resolution, especially at low counts. Clinical trials would be required to assess this method of image postprocessing

Line array transmission sources for SPECT attenuation correction: design and reconstruction

Correction for non-uniform attenuation in SPECT generally requires measurements of radiation transmittance through the patient and reconstruction of the data to form an attenuation image, or mu-map. For nuclear cardiac studies it useful if the emission and transmission data for each projection view can be acquired simultaneously using non-overlapping energy windows. This simplifies the registration of the emission and transmission data. Large area transmission sources are desirable to avoid data truncation; however, 2D-planar liquid sources are cumbersome and extended solid area sources of Gd-153 or Am-247 are impractical. Co-57 sheet sources present spectral overlap problems for imaging of Tc-99m tracers. With Gd-153 line arrays, one can achieve the benefits of 2D-planar sources, low truncation and simultaneous emission/transmission measurements, using lightweight static mechanical attachments to the SPECT camera system. A new method is proposed to determine optimal positions for the lines of the transmission array based on maximizing the entropy of the transmitted flux through the patient. Transmission reconstruction using parallel beam filtered back-projection yields attenuation maps with poor spatial resolution and significant aliasing effects. The degradations of image quality become worse as the angular separations of the lines as seen by the detector increase. To improve the reconstruction of line array transmission data a maximum likelihood modified gradient algorithm was derived. The algorithm takes into account emission-to-transmission down scatter as well as the overlapping of radiation patterns of the individual lines. Ordered subset versions of algorithms are explored. Image quality is assessed with simulations based on an attenuation map derived from CT.

Enhanced Feature Extraction in Planar Nuclear Medicine Using Pixon® Minimum-Complexity Image Processing

This paper introduces a new feature-extraction-receiver-operating-characteristic (FEROC) test to evaluate the ability of image processing to improve the accuracy of the diagnostic information extracted from medical images. The test is applied to simulated planar nuclear images processed by the Pixon minimum-complexity method, originally developed in astronomy, which adaptively smoothes the images to bring out subtle contrasts with minimal loss of resolution. In addition, the processed images are fused with the raw counts with varying blending ratios to allow the readers to detect features with low signal-to-noise ratio, which the Pixon processor might smooth over because of their low statistical significance. The major conclusions from the study are: (1) Pixon processing can substantially increase the negative predictive value (specificity), i.e., reduce the false-positive rate, (2) the positive predictive value (sensitivity) increases modestly due to processing for Pixon blending <50% but declines to ~70% for higher blending percentages (unless it was already below 70% for the raw counts, in which case it is unaffected), (3) the improvement in the negative predictive value is most significant for count levels at which it is already ~80-90% for the raw counts (typical of clinical values), and (4) the improvement is less significant at lower counts (for which the negative predictive value is usually below clinical values) or higher counts (where there is not much to improve).

Analysis of the impact of CT based attenuation correction, scatter compensation and 3D-beam modeling on SPECT image quality

Non uniform photon attenuation and Compton scatter degrade nuclear medicine SPECT images by removing true counts or adding unwanted counts, respectively. For quantitative SPECT, iterative reconstruction algorithms, such as OSEM, allow 3D collimator modeling and compensation of the effects of scatter and attenuation. In this work, we investigate, through ROI and SPM analysis of phantom and computer simulation studies, how a combination of compensation strategies affects image quality and quantitative accuracy. Phantoms were imaged with a SPECT and a CT scanner. With these acquisitions, semi-quantitative analysis was performed for the following reconstruction strategies: OSEM-3D without attenuation and scatter compensation; OSEM-3D CTAC (with attenuation correction); and OSEM-3D with attenuation and scatter compensation (CTACSC). For the simulated dataset, the SNR was 50.5 with CTACSC compared to 27.5 without any compensation and OSEM-3D CTACSC produced reconstructed images with contrast within 0.23% of the true image with a standard error of 21 counts. Without compensation, the error increases to 2382 counts. The implementation and design of the OSEM-3D CTACSC approach proved effective, with improved visual quality and quantitative accuracy of the SPECT images.