PET Pulse Timing Estimation via Curve Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current PET systems face challenges in cost, efficiency, and accuracy due to limitations in detecting high-energy photons and determining the start time of electronic pulses, which affects the precision of coincidence pairs and overall image quality.
Innovation Solution
A method for estimating the start time of electronic pulses using parameterized ideal curve shapes, normalization of pulse signals, and a reverse lookup table, combined with the use of FPGAs for processing and filtering, reduces noise and increases timing resolution without the need for per-channel ASICs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If per-channel ASICs are used for timing measurement, then timing precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent merges the timing measurement function into a centralized digital processing unit that handles multiple detector channels, eliminating the need for separate per-channel ASICs. The digital timing unit processes signals from all channels using a unified approach, reducing overall device complexity while maintaining timing precision through software-based algorithms.
Solution Approach 2:
The patent replaces hardware-based timing measurement (ASICs) with a digital/software-based timing measurement system. By using digital signal processing and software algorithms to determine pulse start times, the system achieves comparable or superior timing precision without requiring complex dedicated hardware for each channel.
2Measurement precision
If per-channel ASICs are used for timing measurement, then timing precision is improved, but manufacturing cost increases
Solution Approach 1:
The patent consolidates timing measurement functionality into a shared digital processing resource that serves multiple detector channels. This merging approach reduces the total component count and manufacturing complexity, leading to lower production costs while maintaining high timing precision through efficient digital algorithms.
Solution Approach 2:
The patent employs standard, commercially available digital processing components and FPGAs instead of expensive custom ASICs. By using off-the-shelf digital logic and programmable devices, the system achieves the required timing precision at a fraction of the cost of dedicated hardware implementations.
3Device complexity
If traditional pulse timing methods are used, then device complexity is reduced, but timing resolution accuracy deteriorates
Solution Approach 1:
The patent applies preliminary digital filtering and signal conditioning to pulse signals before timing measurement. By pre-processing the signals to enhance their characteristics and reduce noise, the system achieves high timing resolution accuracy without requiring overly complex measurement hardware.
Solution Approach 2:
The patent uses feedback mechanisms where the digital processing unit continuously refines timing measurements based on observed signal patterns and statistical analysis. This iterative approach allows the system to achieve high timing precision through software-based correction and optimization rather than complex hardware design.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A method for estimating the start time of an electronic pulse generated in response to a detected event, for example the start time for pulses received in response to photon detection in positron emission tomography, includes providing a detector that detects an external event and generates an electronic analog pulse signal. A parameterized ideal curve shape is selected to represent analog pulse signals generated by the detector. Upon receiving an analog pulse signal, it may be filtered, and then digitized, and normalized based on the area of the digital signal. Using at least one point of the normalized digital pulse signal, a curve from the parameterized ideal curve shape is selected, that best represents the received analog pulse signal, and the selected curve is used to estimate the pulse start time.