PEPconnect

PET Physics and Instrumentation

This e.learning will introduce the viewers to the concepts of radioactive decay, tissue interaction and coincidence detection in PET/CT imaging. Also basic information regarding the hardware and software necessary to produce a diagnostic images will be discussed.

Welcome to the PET/CT Physics and Instrumentation web based training. This tutorial will take you through the basics of PET and CT physics as well as the instrumentation of your PET/CT scanner. You should now be able to: Define the basics of radioactive decay and interaction in tissues and materials. State the concepts of coincidence detection and corrections that are applied to this. Describe the theory of PET image reconstruction. State the performance characteristics of PET imaging. Describe CT physics and instrumentation. List the basic concepts of PET instrumentation Describe how CT and PET systems are integrated In this section, we will discuss radioactive decay and it’s interaction with matter. Nucleus Protons – charge of +1 Neutrons – charge of 0 Electrons – charge of -1 Unstable Unsuitable composition of neutrons, protons or excess energy. Emission of radiation Heavy nuclei Discrete energy Very short travel range Energy range – 3-8 MeV No use in Nuclear Medicine Neutron rich radionuclides Neutron is converted to a proton Emission of beta particle and anti-neutrino Negatively charged beta particles interact with positively charged nucleus X-rays are released Proton rich Emission of a positron along with a neutrino Proton is converted to a neutron Transition energy of 1.022 MeV Proton rich Energy less than 1.022 MeV Electron from the nearest shell to the nucleus is captured by proton in the nucleus to produce a neutron. Nucleus has excess of energy above the ground state Exists in an excited energy state Lifetime of excited energy state expires Nucleus will eventually reach ground state Energy differences appear as gamma photons In this section, we will discuss the different ways that radioactivity can interact with matter. Alpha, beta and positrons Pass through matter Lose energy by interacting with orbital electrons Positron Interacts with electron and comes to a rest Positron and electron are annihilated Produce two 511 keV photons Emitted 180 degrees Detection of gamma photons in coincidence is basis of PET imaging Gamma passes through absorber Transfers entire energy to an inner shell electron Electron is ejected from inner shell Vacancy is filled by electron from upper shell Results in an emission from an x-ray Higher energy gamma interacts with an outer shell electron Some energy from the gamma ray is transferred to outer shell electron Electron is ejected. Lower energy gamma ray will appear as scattered photon and a Compton electron Gamma ray energy is higher than 1.022 MeV Photon interacts with nucleus of the absorber atom during it’s passage through it. Produces positron and electron The energy is shared between the positron and the electron When radiation passes through absorber, the following occur: Photoelectric Compton Pair Production Pass through Combination of the 3 processes is called attenuation Thickness of absorber which attenuates the initial photon beam intensity to one half. HVL increases with higher energy of the photon HVL decreases with increasing atomic number In this section, we are going to discuss PET coincidence detection and the corrections that need to be applied. Sinogram Coincidence events stored on a PET system 2 dimensional histogram of the LORs in a given plane Data is collected as the integral activity along parallel rays for each angle from 0 to 180 degrees. Projection for the angle 0 at the top of the matrix and the other angles in ascending order. SPECT Each projection image represents data acquired at that projection angle across all slices PET Each sinogram represents the data acquired for a slice across all angles There are 3 types of events that can occur during positron emission. Each is significant and has a direct effect on your image quality. Select the tabs below to view the details of each type. Strikes the detectors at 180 degrees Fall within the coincidence timing window or 4.5 nsecs Considered a “valid” event. Strike the detector at 180 degrees Originate from two different annihilation events Too many of these events reduce image quality. Increased occurrence with the increase in specific activity.   Occur from the same annihilation event Strike the detectors within the same coincidence timing window Do not strike the detector at 180 degrees due to deflection or attenuation In this section, we will discuss ways to reduce and correct some of the challenges faced in PET imaging. Randoms increase by: Increasing the width of the energy window Increasing the coincidence timing window Increasing the specific activity Too many random events add to the background causing artifacts and loss of image contrast. Faster electronics LSO, PICO electronics and HD•PET 4.1 nsec coincidence timing windowo Delayed coincidence timing window Random events are measured as they strike the detectors Delayed coincidence timing window Higher specific activity = More randoms Using prescribed doses and scanning longer decreases random events Related to size and density of the object Locations relative to the detectors Not directly measured Estimates Mathematical models Reduced energy window Range for 511 keV Smaller the window, the more scatter is rejected. Estimation of the scatter physics Watson method of scatter correction µ map is obtained from transmission image Calculations are used to approximate the scatter in the emission image Scatter corrected image Absorption of 511 keV photons by tissue Causes non-uniformities in the images Loss of events from central body tissues compared to peripheral tissues Patient dependent Size, density and thickness of organs will affect amount of attenuation Conventional PET scanning Transmission source rotates around the patient for transmission scan Ge68 rod sources Coefficients from transmission scan used to create a µ map Applied to emission data to create an attenuation correction image CT is used for AC Advantages of CT for AC Increased statistical quality Greater resolution Shorter acquisition time Increase patient comfort Decrease patient motion Time it takes for 511 keV to produce a recorded event During this time, second event is unable to be recorded Reduction Detectors with shorter scintillation decay time Faster electronic components LSO Decay time of 40 ns PICO electronics Improves digital time resolution Faster processing of incoming signals In this section of the tutorial, we are going to explore the different methods of PET Reconstruction. Simplest algorithm Counts along LOR detected by pair of detectors are projected back along the lines from which they originated Process is repeated for all LORs in the sonogram Counts from each subsequent LOR are added the counts of the preceding backprojected data Results are back projected original object Backprojection has “star pattern” artifacts Results in blurring of the object Filter applied – less blurring Filtered projection data is backprojected Quick algorithm Too noisy for PET Preferred method of PET image reconstruction Several methods available Steps for one iteration: Initial estimate of the image created Projections are computed from the estimated image Comparison to the measured projections Discrepancies are calculated Corrections are made Updates are applied to the estimated image Repeat as necessary Widely used Updates image during each iteration Steps: Guess of pixel count density for reconstructed image Numerically projects the data forward into a trial projection dataset Trial projections are compared to measured projections Adjustments are made Requires many iterations Demands long computation time Modification of MLEM OSEM groups the projections into subsets with fixed angle Accelerates the computation process Converges on an image that is comparable to MLEM Higher the number of subsets used, the faster the reconstruction Extension of OSEM Artifacts in areas of high attenuation As LORs pass though areas of high attenuation, the attenuation is weighted. Modest increase in computation time over OSEM In this section, we will explore the different performance characteristics in PET imaging. The ability of a device to faithfully reproduce an object. Clearly depicting variations in distribution of radioactivity Described as minimum distance between two points in an image that can be detected by a scanner. Factors that contribute to spatial resolution Detector size Positron travel range Non-colinearity Reconstruction method Localization of detector Standard Detector (Non-Siemens) 6.3 x 6.3 mm (250 mm3 voxels) Siemens HI-REZ and Ultra HI-REZ 4 x 4 mm (64 mm3 voxels) Smaller detectors Increase spatial resolution Decreases partial volume effect Increases quantification accuracy Improves sensitivity 3 mm lesion detectability Positron Travels a distance in tissue Losing most of it’s energy Annihilated after capturing an electron Site of the positron emission is different than the site of annihilation and detection Distance increases with positron energy Degrades spatial resolution Degrades image quality Each positron emitter has it’s own average travel range in water. High density tissues – decrease range Low density tissues – increase range Arises from the deviation of the two annihilation photons from the exact 180˚ position. Deviation of + or – 0.5˚ Displacement causes decrease in spatial resolution Increases with increasing distance between two detectors Too smooth filter can effect spatial resolution Filter with too high cut off value can introduce noise Errors in localization of XY event analysis with block detectors TrueX Technology Millions of accurately measured point spread functions High definition PET images Photons Center of FOV LOR correctly localized Far from center of FOV Less likely calculated correctly Enters the crystal at an angle Continues to another crystal before scintillating Conventional PET Uses the same reconstruction principles across FOV Does not take into account detector geometry and mispositioning of LORs Fuzzy edges Increase distortion further from center of FOV HD•PET Millions of accurately measured PSFs. Positions LORs in actual geometric location Reduces blurring and distortion in final image Conventional PET Positron annihilation would be registered Position along path unknown Time of Flight Used to augment PET image reconstruction Faster detectors are able to measure difference in arrival time of the two gamma rays Estimates the position of annihilation along the path. Fewer iterations needed and less image noise Defined as number of counts per unit of time detected by the device for each unit of activity present in the source. Counts/Second/Microcurie or cps/µCi Depends on: Geometric efficiency Detection efficiency PHA window settings Dead time 2D Imaging Tungsten septa in field of view to reduce scatter Less scatter, less sensitivity 3D Imaging More scatter, more sensitivity Random variation of pixel counts across an image Reduced by increasing number of counts Longer acquisition times Decreased patient comfort Increasing dose Increased radiation dose Increased random events Deadtime loss Improving detector efficiency Limited to scanner design Main sources of statistical error Randoms Scatter NECR Effective number of counts as a function of activity in the FOV. Maximizing NECR minimizes image noice PICO Electronics Improves the Biograph NECR   In this section, we are going to explore Basic CT Physics and Instrumentation. Atoms composed of: Nucleus Protons Neutrons Electrons Bremsstralung X-ray Production in X-ray tube. Wilhelm Conrad Roentgen Glass tube under high vacuum Electrons accelerated at very high speeds towards a target made of tungsten High atomic number Many electrons High melting point Predominate radiation in X-rays Braking or decelerating radiation Electron coming from the atom is slowed down 2 types of interaction Incoming electron collides with the nucleus of the atom Electron nearly collides with the nucleus Energy released in form of a photon X-ray beams are comprised of many photons traveling at the speed of light. Fast moving electrons may collide with an electron target 2 steps: Electron in the target is knocked out of the atom. Electron has a tendency for another target to move into the vacancy Produces excess energy Results in x-ray photon. Very little produced in x-ray tube Patient’s tissues are “cut” into sections First used in conventional x-ray imaging X-ray tube would move in an arc around area of interest X-ray source and film needed to be moved into next position and exposure would be repeated. Time consuming Poor image quality Computers provided breakthrough Used to mathematically reconstruct the images Produced slices Vacuum glass housing Anode Made of tungsten Copper rod for heat dissipation Housed in a glass envelope that is a vacuum High speed electrons hit the anode and x-rays are produced. Cathode Emits electrons that move toward the target Thermionic emission Electrons are accelerated by high voltage applied to the tube Anode is the target Mounting on the rotating part of the gantry Siemens exclusive tube Available on Biograph 20, 40, 64 and Edge systems Direct anode cooling Conventional tube – 120 secs cooling Straton tube – 20 secs cooling Tube assembly is compact Allows for faster rotation speed Withstands higher G forces Reduces size of path of photons emitted from the x-ray tube No collimation would cause: Unnecessary radiation Scatter Pre-patient collimation Restricts the beam immediately after exiting the CT tube, before entering the patient. Thick metal plates Can be adjusted Determines slice thickness in single row detectors Last stop for x-ray photons Photons have to meet the following to be measured. Enter the detector Collide with a detector atom Produce a measurable event Single row or array of the detectors along a curved arc opposite the x-ray tube Collimated into a thin beam Beam is directed through a slice of interest in the patient’s body Determines slice thickness on single slice system Contains multiple parallel rows of detector elements Arranged in curved arcs opposite the x-ray tube. Allows system to acquire more than one slice during rotation X-ray beam is collimated to be thick. Beam is divided into total number of slices acquired in a single rotation. Determines acquired slice thickness Final thickness is determined during reconstruction Slice Collimation Collimated by the tube collimator determining the Z coverage rotation The thicker the slice collimation, the more coverage per rotation you are going to get. Slice Thickness Collimation size of the tube collimator divided by the number of active detector channels. In this section, we are going to discuss the basics of PET scanner instrumentation. High density Efficient scintillator Emit large amount of light Light emitted quickly Characteristics of the detector. Stopping power of detector for 511 keV photons Scintillation decay time Light output per keV of photon energy Energy resolution of the detector 13 x 13 matrix LSO crystal Coupled to 4 PMTs PMTs convert light photons into an electrical pulse. PMT is vacuum glass tube containing: Photocathode at one end 10 dynodes in the middle Anode at other end Select the hotspots in the image below to view PET Scanner Instrumentation locations. Corrections Physical Mathematical 2D Tungsten septa in FOV Restrict the acceptance angle of the coincidence event Decreases sensitivity Longer scan times Fixed or retractable 3D Imaging No septa in FOV Sensitivity increased 4 to 6 times over 2D Increase in scatter and randoms Mathematically corrected In this section, we will discuss how PET and CT have integrated into a hybrid modality. Functional and anatomical imaging modalities developed independently. Late 90s and early 2000s: CT was a single slice modality PET and SPECT were volume modalities Reconstruction techniques of both modalities were based on common principles. Became important in the 1980s First attempts were software based Successful in the brain because it is a rigid organ fixed in the skull. Other parts of the body, rigid fusion was problematic. Dual modality system was needed Respiration Shift of internal organs Interscan delay Patient positioning Bed curvature Eliminates the need for a separate lengthy PET transmission scan. Noiseless attenuation correction factors CT keV must be scaled the PET keV Hybrid scaling algorithm was developed using mass attenuation coefficient for all tissues Patient respiration Use tidal breathing Respiratory gated acquisition Truncation of CT FOV CT reconstructed FOV – 50 cm Aperature is 78 FoV HD FoV allows the entire FOV to be used for AC IV or Oral Contrast CTs Contrast High atomic number substance Increases attenuation in vessels, bowel and other structures above normal values Up to 700 HU can be achieved. At 300 HU Algorithm could incorrectly segment and scale structure as bone Potential artifacts in corrected images Compare corrected images with uncorrected images Injection of a radioactive tracer Uptake 30-60 mins Patient is positioned Topogram acquired For FOV localization CT Scan Shallow breathing Automatically reconstructed Scatter and attenuation correction PET PET Scan Whole body scan – pelvis to head Depending on the scanner Continuous bed motion Step and Shoot Bed positions - 2 mins/bed or more Reconstruction for each bed position performed after complete.   In this last section, we are going to review the types of computer systems that will come with your Biograph mCT and True Point Scanners. Acquisition Reconstruction Registration Scheduling Protocol Selection Standard Application Evaluation Data Transfer Communicates with the scanner Calculates the images for each CT slice Passes data to ICS Does not require direct operation Controlled by the ICS syngo MI Workplace Optional satellite console Reconstruction 3D Evaluation PET/CT Image Fusion PET Data Acquisitions Temporary storage of PET raw data PET gantry service procedures Reconstructs PET Sinograms into image Final storage location for PET list mode data Conventional System Connects 12 block detectors connected to 12 DEAs 3 rings of detectors TrueV Connects 16 block detectors to 12 DEAs 4 rings of detectors Select the highlighted areas in the image below to see a typical PET/CT floorplan. Imagine sonogram as a pie Slices of the pie are subsets Each pie is sampled by the number of iterations chosen. The more iterations, the sharper the images By the end of this training you will be able to:   Define the basics of radioactive decay and interaction in tissues and materials State the concepts of coincidence detection and corrections that are applied to this Describe the theory of PET image reconstruction State the performance characteristics of PET imaging Describe CT physics and instrumentation List the basic concepts of PET instrumentation Describe how CT and PET systems are integrated

  • biograph
  • truepoint
  • mct
  • excel
  • syngo.via