Discussion and Concluding Remarks

This chapter presented an overview of quantitative PET imaging, including the basic principles and instrumentation, methods of image reconstruction from projections, and some specific correction factors necessary to achieve quantitative images. Techniques for absolute physiologic parameters estimation based on the framework of tracer kinetic modeling are also introduced. Recent advances in instrumentation and detector technologies have partially resolved some of the challenges. Research is still under way to develop new detector materials not only for improving the spatial resolution but also the system sensitivity, while keeping the cost of the instrument acceptable. Another active research area is image reconstruction. Although the filtered backprojection method is still widely used nowadays, it is not an optimal reconstruction technique. Iterative algorithms have been shown to improve the signal-to-noise ratio of the reconstructed images and provide more accurate image reconstruction for low count studies, but the computational complexity and appreciably long reconstruction times as compared with filtered backprojection, as well as the requirement of using some ad hoc techniques to control the visual quality of the reconstructed images, remain a substantial obstacle for routine implementation. It should be noted, however, that a good reconstructed image is not only dependent on the reconstruction algorithm, but also dependent on the sufficient axial and angular sampling of projection data such that the reconstruction artifacts can be minimized. In order to achieve accurate absolute or relative quantification, appropriate data corrections have to be applied prior to image reconstruction.

The use of SPECT for studying physiologic functions deserves mention here. SPECT is another form of emission computed tomography which had its beginning in the early 1960s, initiated by the work of Kuhl and Edwards on transverse-and longitudinal-section scanning with single-photon-emitting radiotracers [15], and the work of Anger on the development of scintillation camera (also known as gamma camera or Anger camera) coupled with photomultiplier tubes [108]. Although the innovative idea of how gamma rays interact with inorganic crystal (scintillator) to produce scintillation light and how optical coupling with pho-tomultiplier tubes helps amplify the scintillation signal may shed some light on detector design in PET, the development of SPECT imaging, however, has been overshadowed by PET for reasons to be detailed. The principles for detecting the emerging radiation and the formation of tomographic images are similar for both PET and SPECT but the underlying physics and the instruments employed are completely different. In SPECT, nuclear decay results in the emission of one or a few uncorrelated photons in the 100-200 keV energy range. A lead collimator drilled with small holes is used to mechanically collimate the incoming photons by allowing those traveling in one particular direction to interact with the scintillator, while all others are absorbed. Rotating gamma cameras with single or multiple crystal detectors are used to form a tomographic image. Here lies the sensitivity differences between PET and SPECT, and partly explains why PET has received much more attention than SPECT for in vivo assessment and quantification of physiologic functions in the body.

Despite the fact that both PET and SPECT suffer from attenuation and Comp-ton scattering of the photons inside the body which can result in image artifacts and loss of quantitative accuracy, SPECT has been largely considered to be nonquantitative and limited to providing qualitative or relative functional images. This is because correction of attenuation and scatter in SPECT are not easy as compared to PET, where attenuation correction is routine (with the exception of whole-body PET). In addition, the spatial resolution of SPECT is inferior to that in PET. Even with triple-headed gamma cameras, the resolution is approximately 8-10 mm FWHM but the theoretical limit of 1-2 mm FWHM can be achieved for PET with new generation of detector technology. Further, typically higher signal-to-noise ratio and lower scatter with PET also helped establish PET as the favorable method for quantitative measurements of physiological parameters.

Although PET will continue to provide insights into biochemical and physiological processes in vivo, access to PET is limited due to the requirement of a cyclotron and high operation costs. Recent advances in quantitative SPECT and the widespread application of multidetector SPECT systems with improved sensitivity and dynamic imaging capabilities have made absolute physiological parameter estimation possible with the much more widely available SPECT. One of the major applications of dynamic SPECT is to quantify myocardial perfusion, which is important for the diagnosis and clinical management of patients with coronary artery disease where a perfusion defect after an intervention may indicate incomplete reperfusion or persistent coronary occlusion. Similar to dynamic PET, compartmental modeling is used in dynamic SPECT to quantify physiologic parameters of interest. It has been demonstrated that myocardial blood flow can be studied by dynamic SPECT imaging of 99mTc-teboroxime [109,110]. Another interesting application of dynamic cardiac SPECT is the measurement of perfusion and the distribution volume in the heart using 201Tl [111, 112]. Other applications include brain imaging with 99mTc and 123I flow agents and some neuroreceptor studies, such as [123I]iomazenil for benzodiazepine receptors [115], [123I]iododexetimide for cholinergic muscarinic neuroreceptors [114], and [123I]iodo-A-85380 for central neuronal nicotinic acetylcholine receptors [115,116], etc., both in human and nonhuman primates.

While PET and SPECT allow absolute measurements of radioactivity concentration in tissue and have the capability to relate the in vivo measurements to physiological functions, there are a number of limiting factors which can impact on their ability to produce reliable physiologic parameters. In particular, the small tissue volume, limitation on the total amount of activity that can be administered, physical artifacts, and low sensitivity (particularly for SPECT) result in measurements with high noise levels. The measurement noise (and mechanical rotation of the detectors in the case of SPECT) has imposed an upper limit on the fastest sampling rate which is >1 sec/sample for PET and >5 sec/sample for SPECT. Furthermore, the short half-life of the radiopharmaceuticals used for PET and SPECT and the need for the patient to remain still on the scanner bed usually limit duration of the PET and SPECT experiments to relatively short periods of time, which vary from minutes to hours rather than days. Slow physiologic processes are therefore difficult to be reliably estimated with both PET and SPECT. All these factors limit the number of parameters which can be reliably estimated from PET or SPECT measurement and complicated compartmental models used for analysis must be simplified. Huang et al. provide detailed guidelines on kinetic model development for PET applications and these principles are equally applicable to SPECT tracer studies.

One intrinsic limitation of PET imaging is the presence of physiological sites of tracer accumulation (secretion or excretion). It is sometimes difficult to differentiate real pathology from an unusual pattern of physiologic accumulation. In contrast, some organs of the body, for instance, the pancreas, uterus, and ovary, do not have significant physiological uptake. Spatially differentiation among these organs is sometimes difficult, particularly when they are closely related to one another. In this respect, structural imaging modalities such as X-ray CT can serve as an excellent anatomical roadmap for the functional PET images.

Integration of anatomical and functional images began to emerge during the late 1980s using software-based registration techniques [117,118], although it was pointed out much earlier by Wagner [119] that this would be the trend for clinical imaging. These software-based registration techniques are applied to register images obtained with different imaging modalities or tracers (in case of registering emission tomographic images), of the same subject at different times. Registration techniques are successful for a rigid organ, such as the brain but they have been found to be problematic for other parts of the body.

Recognizing the advantages of combining the information provided by anatomic imaging and functional imaging, a prototype of an integrated scanner for PET and CT was designed in the early 1990s and a commercial hybrid scanner, named "PET/CT," has been developed recently [120]. PET/CT is a completely new imaging technique that will likely revolutionize the conventional habit of acquiring and reading PET and CT data separately in the clinical environment. PET/CT imaging will enhance the combined utilization rate of what used to be PET-only or CT-only imaging as it provides, simultaneously, co-registered (fusion) images of both functional and anatomical information in a single acquisition. A potential advantage is the use of CT images for attenuation correction of PET emission data, thereby the need for acquiring a separate, lengthy transmission scan can be completely eliminated. Figure 2.13 shows a combined PET/CT

Figure 2.13: (Color slide) Combined PET/CT scan on a 72-year-old woman with a primary pancreatic cancer. From left to right: FDG-PET image, CT image, and fused PET and CT images. The scan shows abnormal FDG uptake in the pancreas (arrow). The fused image shows good alignment of two modalities and enables uptake to be localized to pancreas (arrow).

Figure 2.13: (Color slide) Combined PET/CT scan on a 72-year-old woman with a primary pancreatic cancer. From left to right: FDG-PET image, CT image, and fused PET and CT images. The scan shows abnormal FDG uptake in the pancreas (arrow). The fused image shows good alignment of two modalities and enables uptake to be localized to pancreas (arrow).

scan performed on a 72-year-old woman for investigation of unknown primary malignancy, and the combined PET/CT revealed a primary pancreatic cancer. Although the PET/CT imaging is still in its infancy and a number of technical problems remain to be solved, it is anticipated that the combination has the potential to solve many of the present diagnostic challenges associated with whole-body oncologic imaging and has an important role in surgical planning, radiation therapy, treatment monitoring, and diagnosis of disease.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment