Tangential TOF Mask for PET Random Coincidence Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In positron emission tomography (PET) systems, random coincidences due to finite coincidence windows introduce substantial errors in image reconstruction, and existing methods to reduce random events often compromise true coincidence detection rates.
Innovation Solution
A method utilizing time-of-flight (TOF) resolution and acquisition field of view (FOV) information to create a tangential TOF mask that filters out random coincidences outside the reconstruction FOV, maintaining true events, by applying the inequality c² * (ta - tb) * cos(θ) ≤ d² / 4 - s² + n * σTOF, where ta and tb are event arrival times, θ is the axial tilt angle, d is the FOV diameter, σTOF is the timing resolution standard deviation, and n is a confidence interval factor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fixed coincidence window method is used to detect true coincidences, then true coincidence detection is maintained, but random coincidences comprise a large portion of recorded events (30%-50% random fraction)
Solution Approach 1:
The patent segments the coincidence detection process by introducing a tangential TOF mask that divides the detection space into valid regions (within FOV) and invalid regions (outside FOV). This segmentation allows the system to maintain the fixed coincidence window for true event detection while filtering out random coincidences that fall outside the masked region, thereby resolving the contradiction between maintaining true coincidence detection and reducing random coincidences.
Solution Approach 2:
The patent applies local quality by creating a spatially varying acceptance criterion through the tangential TOF mask. Different regions of the detection space have different acceptance properties: regions within the FOV maintain full acceptance for both true and random events, while regions outside the FOV reject random events. This local differentiation allows the system to reduce overall random coincidences while preserving true coincidence detection in the relevant imaging region.
2Object-generated harmful factors
If a truncated time window is used to reduce random events, then random coincidences are reduced, but true coincidence detection rate also decreases
Solution Approach 1:
The patent transitions from a one-dimensional time-based filtering approach (truncated time window) to a two-dimensional spatial-temporal approach by introducing the tangential TOF mask in the spatial domain. This dimensional change allows the system to filter random coincidences based on their spatial origin rather than solely on timing, thereby reducing random events without compromising true coincidence detection rates.
3Object-generated harmful factors
If CT morphological information is used to tailor the time coincidence window, then random events are reduced, but about 10%-15% of true coincidences are rejected
Solution Approach 1:
The patent introduces the tangential TOF mask as an intermediary filtering element that operates independently of patient-specific CT morphological information. This intermediary structure provides a generic yet effective means of reducing random coincidences based on the known FOV geometry, avoiding the need for patient-specific customization that would otherwise lead to rejection of true coincidences at tissue boundaries.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces random events by 50% while increasing the noise-equivalent-count rate by 12% without affecting true coincidence detection, as demonstrated with an IEC phantom, and can be applied to various FOV shapes conforming to patient cross-sections.
Implementation Method 1
In time-of-flight (TOF) imaging, the time within the coincidence interval at which each gamma photon in the coincident pair is detected is also measured. The time-of-flight information provides an indication of the annihilation location of the detected event along the line of coincidence.
Implementation Method 2
The inequality c² * (ta - tb) * cos(θ) ≤ d² / 4 - s² + n * σTOF, where ta and tb are event arrival times, θ is the axial tilt angle, d is the FOV diameter, σTOF is the timing resolution standard deviation, and n is a confidence interval factor.
Data Source
AI summary
A method and apparatus for reducing random events in positron emission tomography (PET) list mode data, the method including the steps of obtaining, for a PET scanner having a given reconstruction field of view (FOV), time-of-flight (TOF) prompt list-mode count data that includes TOF information, the TOF prompt list-mode count data including a plurality of entries; and filtering the obtained prompt list-mode count data by removing those entries in the obtained prompt list-mode count data that represent emission points lying outside a tangential TOF mask to obtain filtered list-mode count data.