- Positron Emission: PET imaging relies on the detection of gamma rays emitted from a positron-emitting radionuclide. When a positron meets an electron, annihilation produces two 511 keV photons traveling in opposite directions. The scanner detects these coincident photons to localize the source.
- Coincidence Detection: Two detectors opposite each other register a coincidence event if both photons arrive within a narrow time window, typically 4-6 nanoseconds. This electronic collimation eliminates the need for physical collimators, improving sensitivity compared to SPECT.
- Reconstruction Algorithms: Raw coincidence data must be reconstructed into tomographic images using algorithms like filtered backprojection or iterative methods. Modern scanners use iterative reconstruction with corrections for attenuation, scatter, and random coincidences.
- Attenuation Correction: Tissue attenuates gamma rays, so correction factors derived from a transmission scan (or CT in PET/CT) are applied to the emission data. Without correction, deep structures would appear less radioactive.
- Scatter Correction: Some photons scatter before detection, losing energy and direction. Scatter correction algorithms estimate and subtract the contribution of scattered events to improve image contrast and quantitation.
- Normalization: Detector efficiency varies; normalization factors are applied to equalize response across all detector pairs. This ensures uniform sensitivity and consistent image quality.
- Data Binning: Coincidence events are sorted into sinograms, which are matrices representing projection data at various angles and offsets. Each sinogram row corresponds to a specific line of response.
- Time-of-Flight (TOF) PET: Some scanners measure the time difference between photon arrivals to estimate the annihilation location along the line of response. TOF improves signal-to-noise ratio and reduces reconstruction time.
- Random Coincidence Correction: Random coincidences occur when two unrelated photons from separate annihilations are detected within the coincidence window. These events add background noise and degrade image contrast. Correction methods estimate randoms using a delayed coincidence window or singles rate measurements and subtract them from the prompt coincidences.
- Energy Discrimination: Detectors measure the energy of each incoming photon. A typical energy window around the 511 keV photopeak, such as 350-650 keV, is applied to reject scattered photons that have lost energy. This improves image quality by excluding events that would otherwise be mispositioned. The energy resolution of the detector material directly affects the efficacy of this discrimination.
- System Dead Time: At high count rates, the scanner's electronics require a finite time to process each event, during which additional events may be lost. Dead time correction models this loss and applies a multiplicative factor to the measured counts. Proper correction is critical for quantitative accuracy, especially in dynamic or high-activity studies. Each scanner's dead time behavior is characterized during calibration.
Role: PET Imaging Operations Specialist
- Operating the PET scanner according to standard protocols
- Applying quality control procedures daily
- Monitoring patient safety during scanning
- Documenting operational parameters and patient data
- Interpreting diagnostic images or making clinical diagnoses
- Administering radiopharmaceuticals (performed by authorized nuclear medicine technologists)
- Modifying scanner hardware or software without vendor approval
- Performing radiopharmaceutical preparation (handled by radiochemists)
- Annihilation
- The process where a positron and electron combine, producing two 511 keV photons.
- Coincidence Event
- Detection of two photons within a time window, indicating annihilation.
- Sinogram
- A data matrix used in reconstruction, where rows represent projection angles.
- PET imaging uses positron-emitting radionuclides and coincidence detection.
- Corrections for attenuation, scatter, and normalization are essential for accurate images.
- Time-of-flight technology improves image quality.
- Operators focus on protocol execution and quality control, not diagnosis.