CT Functional Imaging Using Simulated PET Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
PET-CT imaging is limited by patient motion, high costs, radiation exposure, local complications from radiotracers, and allergies, making it undesirable for clinical use.
Innovation Solution
Identify functional features from CT images using radiomic analysis, generative machine learning models, and generative adversarial networks to generate simulated functional images without the need for radiotracers, allowing for the visualization of functional activity directly from CT scans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If PET-CT imaging is performed to identify functional features, then functional activity detection capability is improved, but patient motion between PET and CT studies increases causing misregistration
Solution Approach 1:
The patent combines PET and CT imaging into a single integrated PET-CT system that acquires both functional and anatomical images simultaneously or in close succession, eliminating the temporal gap that causes patient motion between separate studies. This merging approach maintains functional detection capability while ensuring accurate co-registration by reducing the time window for motion artifacts.
2Measurement precision
If radiotracers are administered for functional imaging, then functional feature identification is improved, but production cost and facility requirements increase
Solution Approach 1:
The patent creates a simulated PET image that copies the functional information typically obtained from radiotracer-based PET imaging, but derives it instead from standard CT images through advanced image processing and transfer learning techniques. This copying approach replicates the functional imaging capability without requiring expensive radiotracers or cyclotron facilities, making functional imaging accessible in常规CT settings.
3Measurement precision
If radiotracers are injected into patients for PET imaging, then functional imaging capability is improved, but risk of local complications and allergies increases
Solution Approach 1:
The patent extracts the essential functional imaging capability from the radiotracer injection process itself, separating the functional information acquisition from the harmful injection step. By using transfer learning to map anatomical features from CT to functional features typically seen in PET, the method removes the need for radiotracer administration while preserving the functional imaging benefit, thereby eliminating injection-related complications and allergies.
4Measurement precision
If CT scan duration is extended to improve functional analysis, then functional feature detection is improved, but patient exposure to radiation increases
Solution Approach 1:
The patent makes the standard CT scan serve multiple functions: it provides both the常规anatomical imaging function and, through transfer learning and simulated PET generation, also provides functional imaging capability. This multi-functionality allows functional feature detection to be achieved using the same CT radiation dose that would normally be used for anatomical imaging alone, avoiding additional radiation exposure while expanding diagnostic capability.
Data Source
AI summary
Methods, apparatus and computer readable media are provided for identifying functional features from a computed tomography (CT) image. The CT image may be a contrast-enhanced CT image or a non-contrast CT image. According to some examples, methods, apparatus and computer readable media are also provided for using machine learning to identify functional features from CT images. According to some examples, simulated functional image datasets such as simulated PET images or simulated SUV images are generated from a received CT image.


