As discussed in a previous blog post ”Small Animal Imaging Systems” about the challenges in disease model studies, treatment development, and other research requiring preclinical activities, the use of small animal imaging technology and equipment is becoming crucial to ensuring productivity, accelerating development and enhancing research
This specialized equipment, such as positron emission tomography (PET) systems, allows examination of the anatomical, physiological and functional parameters at the molecular and cellular levels in living organisms, making it possible to assess in depth the impact of the research subject (for example, a new drug).
Although sophisticated, and providing interesting results, the conventional PET system design presents some limitations. These limitations could be improved using new technologies such as GAPD array detectors and free-running ADC- and FPGA-based signal processing methods, such as those provided by Nutaq FPGA data acquisition systems. These limitations and possible improvements are addressed in this series of two blogs posts.
This first post provides an overview of the conventional design of PET and MicroPET systems. We’ll look at the PET system’s components and processing methods to get a better understanding of how these systems function, and then we’ll present an overview of some limitations inherent in this conventional design.
PET/MicroPET Design And Operation Overview
The PET/MicroPET is a three-dimensional nuclear imaging system that allows the assessment of functional processes in live humans or animals. The PET imaging system detects pairs of gamma rays emitted indirectly by a positron-emitting radionuclide (tracer) injected into the subject and then processes detected rays to provide images of internal structures of the body [Reference: Wikipedia].
A PET/MicroPET scanner is composed of three main parts: the sensor (or detector) unit, the coincidence processing unit, and the image reconstruction unit. See here and here for schematic diagrams of a MicroPET scanner, a photomultiplier tube, and scintillator block design images.
The PET Sensor
The ring sensor unit is generally built of many scintillator blocks, and each of these blocks is optically connected to four 4x photomultiplier tubes (PMT). In brief, the emitted photons which are incident on the scintillator create energetic electrons. As the electrons pass through the scintillator, they excite other electrons, which give off light as they decay back to their ground state. The light is then captured by the four PMTs. When the photons collide with the photocathode of the PMT, photo-electrons are emitted. These electrons then travel through the PMT and hit dynodes which release more and more electrons in the PMT. Finally, these electrons collide with the anode and generate electric current pulses. [Reference: University of Illinois and University of Washington]
The Coincidence Processing Unit
These electric current pulses are then transmitted and processed by the coincidence processing unit (such as our Nutaq FPGA acquisition system). As we remember, the pair of photons, produced when the radioactive positron annihilates the electrons in the subject body, is emitted at opposite direction (approx. 180°). The coincidence processing unit then tries to identify and gather the two electric pulses associated to pair of photons and to reject scattered photons. This is performed by passing the pulse through, first, a constant fraction discriminator (CFD), which generates a digital pulse when the signal reaches a constant fraction of the peak pulse height and, second, a Time-to-digital converter to evaluate the timing coincidence of the two pulses. Once a pair of pulses is identified, a coincidence event is triggered and a line of where the positron could have been emitted can then be extrapolated. [Reference: University of Illinois]
The Image Reconstruction Unit
The image reconstruction unit processes the coincidence events statistics to reconstruct the image. Even if different algorithms exist, the idea is to create a map of radioactivity as a function of the emission location. An image of the tissues, in which the molecular tracer has become concentrated, could be then reconstructed and analyzed by researchers. [Reference: Wikipedia]
PET Conventional Design Limitations
The Photomultiplier Tube
In a conventional PET system such as described in the previous section, the photomultiplifier tube (PMT) and coincidence processing unit, using a constant fraction discriminator and Time-to-Digital converter, present limiting factors that impact the overall performance, size, weight, and cost of the equipment. [Reference: RMD]
In fact the size of the sensors ring is directly related to the configuration sensor block. PMTs are built in a globe vacuum, similarly to old bulb diode or transistor, and a sensor block is made with a scintillator coupled to 4x PMT. Therefore, as the PMTs size is limited by technology itself, it is difficult to scale them down and the bulkiness of PMTs is a limiting factor in clinical as well as small animal PET cameras. [Reference: RMD]
Another really important limitation of PMTs is their sensitivity to magnetic fields. The crystal decoding accuracy can be degraded as the electrons passing through PMTs are deviated by magnetic field. Therefore, PMTs are unsuitable for multimodal PET/MRI scanner or other applications in the presence of magnetic field. [Reference: Harvard Medical School]
The Constant Fraction Discriminator
The use of a constant fraction discriminator (CFD) processing method is an important limitation from a performance perspective. In brief, true pulse coincidence events could be discriminated while scattered photons could be considered as positive events. The [Reference: University of Illinois] discusses this in greater detail.
The Time-To-Digital Converter
Finally, another component of the processing method impacting the PET performances is the constant Time-to-Digital Converter (TDC). As we learned, this component discriminates the timing coincidence of the pair of pulses. In short, the TDC generates a positive event when the two pulses are within its acceptance criteria window. Therefore, if time window “is too small compared to the time resolution of the detection system, true coincidences will be missed.” On the other hand, “if it is too large, more random coincidences will be counted without significant increase in the number of true coincidences.” Therefore, this fixed TDC window could have an important impact on the image resolution if is not correctly configured. [Reference: University of Illinois]
Upcoming Blog post – Sogang University Successfully Addresses PET System Challenges
In summary, we learned that PET systems are mainly built with 3 main units: 1) the ring sensor, 2) the coincidence processing unit, and the Image reconstruction unit.
Moreover, we discussed about size, MRI compatibility, and performance limitations of PET system due to technology used in the conventional design of these different units.
In the second blog in this two-part series, we will describe how Sogang University developed a prototype to address these common technical limitations of PET systems. We will also show how Nutaq’s solution (hardware and software) helped resolve some of these issues and how it has accelerated the design and development of this prototype.