Adaptive PET Acquisition Parameters for Clinical Detectability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing molecular imaging (MI) systems face challenges in efficiently and cost-effectively improving the detectability of clinically-relevant information within images, as altering standard data acquisition protocols often results in increased costs and minimal improvement in image quality.
Innovation Solution
A system that adapts MI data acquisition parameters based on comparisons between synthetic PET images generated from CT images using a trained neural network and actual PET images, allowing for real-time adjustments to enhance image quality and detectability of clinically-relevant information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard data acquisition protocols are used in MI, then acceptable image quality is achieved for most patients, but the detectability of clinically-relevant information is limited
Solution Approach 1:
The system performs preliminary action by acquiring scout images at the beginning of the scan and using a trained neural network to generate synthetic PET images from CT images. These preliminary images are compared to detect differences in tracer uptake and distribution, allowing the system to proactively identify regions requiring enhanced imaging before the full acquisition protocol is executed, thereby improving detectability without universally extending scan time for all patients.
Solution Approach 2:
The system applies dynamics by adaptively adjusting data acquisition parameters based on real-time comparison between scout images and synthetic PET images. The acquisition protocol transitions from a static standard protocol to a dynamic, patient-specific protocol where parameters such as scan time and gating are modified only for regions showing differences in tracer uptake, enabling improved detectability while maintaining time efficiency for standard cases.
2Measurement precision
If data acquisition protocols are changed to improve image quality, then detectability of clinically-relevant information may improve, but costs increase due to increased imaging room time and processing requirements
Solution Approach 1:
The system implements local quality by applying enhanced data acquisition parameters only to specific regions of interest where differences in tracer uptake or distribution are detected through comparison of scout images with synthetic PET images. Instead of uniformly improving image quality across the entire field of view, the system locally adapts acquisition parameters such as scan time and gating for affected regions, thereby improving detectability of clinically-relevant information while minimizing additional imaging room time and processing requirements.
3Loss of energy
If standard acquisition protocols are used, then cost-effectiveness is maintained, but image quality and detectability of clinically-relevant information remain suboptimal for certain patients
Solution Approach 1:
The system employs feedback by continuously comparing scout images with synthetic PET images generated from CT data using a trained neural network. This comparison provides real-time feedback on tracer uptake and distribution patterns, enabling the system to identify patients who would benefit from enhanced acquisition parameters. The feedback loop allows the system to maintain cost-effectiveness for standard cases while automatically triggering quality improvements only when clinically indicated, thereby optimizing the balance between cost and image quality.
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
Enhances the detection of clinically-relevant information in MI images while maintaining cost-effectiveness and time efficiency by dynamically adjusting acquisition parameters to address differences in tracer uptake and distribution.
Implementation Method 1
synthetic PET images generated from CT images using a trained neural network
Data Source
AI summary
Systems and methods include determination of an anatomical image of an object, input of the anatomical image to a trained neural network to generate a synthetic functional image, acquisition of molecular imaging data of the object based on acquisition parameters, reconstruction of a functional image based on the molecular imaging data, determination of a difference between the functional image and the synthetic functional image, change of one of the acquisition parameters based on the difference, acquisition of second molecular imaging data of the object based on the changed acquisition parameters, and reconstruction of a second functional image based on the second molecular imaging data.