Nuclear Imaging Pulse Decoupling for Accurate Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nuclear imaging systems often miss or ignore detection events when multiple events occur near each other in time, leading to inaccurate measurement data and reconstructed medical images.
Innovation Solution
Employing trained machine learning processes to detect and decouple multiple pulses from pileup events, generating accurate energy and position estimates for each pulse using a neural network or deep learning model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional detection methods are used in nuclear imaging systems, then the system is simple and easy to operate, but detection events occurring near each other in time are missed or ignored, reducing measurement precision
Solution Approach 1:
The patent replaces traditional mechanical/electronic signal processing methods with machine learning-based detection algorithms. The machine learning model analyzes pulse patterns and identifies multiple detection events that occur near each other in time, substituting conventional threshold-based detection with intelligent pattern recognition to improve measurement precision without requiring complex hardware modifications
Solution Approach 2:
The patent changes the detection parameters by using time-based pulse pattern analysis instead of simple amplitude thresholding. By examining the temporal characteristics and patterns of signals over time, the system can distinguish between single events and multiple closely-spaced events, improving detection accuracy through parameter transformation rather than adding complex detection hardware
2Productivity
If multiple detection events occur near each other in time, then the system can capture more data, but the measurement data becomes inaccurate due to pileup events
Solution Approach 1:
The patent segments the detection process into multiple stages: first identifying pileup events through machine learning pattern recognition, then separating and analyzing individual pulses within those events. This segmentation allows the system to maintain high data capture rates while improving measurement precision by analyzing each pulse individually rather than treating pileup events as single inaccurate measurements
Solution Approach 2:
The system uses feedback from machine learning models that continuously learn from detection patterns to improve pulse separation accuracy. The model receives input signals, processes them through trained neural networks, and provides feedback about detected pulse patterns, enabling iterative refinement of energy and position estimates for each separated pulse event
Data Source
AI summary
Systems and methods for detecting multiple events during nuclear imaging scans, and for reconstructing images based on the detected events, are disclosed. In some embodiments, an image scanning system scans a subject, and generates a signal characterizing a detection event. The system generates sampled data based on sampling the at least one signal. Further, the system applies a trained machine learning process to the sampled data. Based on the application of the trained machine learning process, the system generates pulse data characterizing a plurality of decoupled pulses. For example, the pulse data may characterize pulse energy values of each of the decoupled pulses, and corresponding times for each of the pulses. Further, the method includes transmitting the pulse data to generate time-coincident pairs for image reconstruction.