Pixelated Detector Noise Correction via Consistency Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nuclear medicine imaging systems, such as SPECT, PET, and CT, face issues with image artifacts and reduced reliability due to malfunctioning pixels in pixelated detectors, which often require the entire detector to be replaced at high cost.
Innovation Solution
A method is developed to identify and dynamically detect noisy pixels in pixelated detectors by determining data consistency metrics and using statistical measures to re-project reconstructed images, allowing for the correction of malfunctioning pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire detector is replaced to address malfunctioning pixels, then image quality and reliability are improved, but cost increases significantly
Solution Approach 1:
The detector is segmented into individual pixels that can be independently evaluated and processed. Malfunctioning pixels are identified and handled separately through statistical analysis and correction algorithms, rather than replacing the entire detector array. This allows selective correction of only the problematic pixels while retaining functional pixels.
Solution Approach 2:
The patent creates a corrected version of the detector data by generating corrected pixel values through statistical measures and image reconstruction algorithms. These corrected values replace the erroneous readings from malfunctioning pixels, effectively creating a copy of the detector output that is free from pixel-level defects without physical replacement.
2Reliability
If the entire detector is replaced to address malfunctioning pixels, then image quality is improved, but time for replacement increases
Solution Approach 1:
The patent performs preliminary evaluation of detector pixels using statistical measures during or before image acquisition. By pre-identifying malfunctioning pixels through consistency metrics and evaluating their impact, the system can prepare correction strategies in advance, enabling rapid software-based correction without time-consuming physical detector replacement.
Solution Approach 2:
The patent substitutes the mechanical process of physical detector replacement with an information-processing approach using statistical algorithms and image reconstruction techniques. Instead of mechanically replacing the entire detector array, the system uses computational methods to identify and correct malfunctioning pixels through data processing and reconstruction algorithms.
3Reliability
If statistical measures are used to identify noisy pixels, then detector reliability is improved, but device complexity increases
Solution Approach 1:
The patent implements self-service evaluation where the detector system automatically evaluates its own pixel performance using statistical measures and consistency metrics. The system performs self-diagnosis and self-correction by identifying malfunctioning pixels through its own acquired data and applying correction algorithms, eliminating the need for external intervention or complex manual calibration procedures.
Solution Approach 2:
The patent changes the parameters of pixel evaluation by introducing statistical measures such as mean, standard deviation, and consistency metrics that quantify pixel performance. By transforming raw pixel data into statistical parameters and using these parameters to identify and correct malfunctioning pixels, the system manages complexity through mathematical formalization rather than additional hardware complexity.
Data Source
AI summary
An apparatus and methods for evaluating the operation of pixelated detectors are provided. The method includes obtaining data values for each of a plurality of pixels of a pixelated detector and determining a data consistency metric for each of the plurality of detector pixels. The method further includes identifying, using the determined data consistency metric, any detector pixels that exceed an acceptance criterion as noisy pixels.