Single-Energy CT Image Transformation via Phase-Specific Deep Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Single-energy CT imaging systems lack the ability to provide increased contrast visualization and are prone to noise and image artifacts, limiting the quality of CT images, especially at lower energy levels.
Innovation Solution
Transforming images obtained at a single peak energy level to different energy levels using a deep learning-based energy transformation model, specifically selecting an energy transformation model based on the contrast phase of the image to achieve improved contrast visualization without reducing image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-energy CT imaging system is used, then the device complexity is reduced, but the contrast visualization quality deteriorates
Solution Approach 1:
The patent creates virtual copies of dual-energy images from single-energy images using deep learning transformation models. The system generates synthetic low-energy and high-energy images that mimic the appearance and diagnostic value of true dual-energy images, allowing single-energy systems to achieve dual-energy image quality without additional hardware complexity
Solution Approach 2:
The patent transforms images between different energy levels by applying energy transformation models that modify the energy parameters of the imaging data. By selecting appropriate transformation models based on contrast phase identification, the system can generate images at different energy levels from a single-energy acquisition, improving contrast visualization without changing the physical imaging system
2Measurement precision
If images are transformed to different energy levels, then contrast visibility is improved, but image quality may be reduced due to noise and artifacts
Solution Approach 1:
The patent performs preliminary identification of contrast phase before applying energy transformation. By pre-classifying the contrast phase (e.g., arterial, venous, delayed) and selecting the appropriate transformation model in advance, the system ensures that the image transformation maintains diagnostic quality and reduces the risk of introducing artifacts or noise
Solution Approach 2:
The patent introduces an intermediate contrast phase classification step that acts as a mediator between the single-energy image and the energy transformation process. This intermediary classification ensures that the correct transformation model is applied, maintaining image quality while achieving the desired contrast enhancement at different energy levels
3Measurement precision
If multiple energy levels are used for imaging, then tissue characterization quality is improved, but the device complexity and scanning time increase
Solution Approach 1:
The patent generates virtual multi-energy images from a single-energy acquisition, creating synthetic copies of what would be obtained from multi-energy scanning. This allows post-processing tissue characterization at multiple energy levels without requiring actual multi-energy acquisitions, thereby maintaining diagnostic quality while reducing scanning time
Solution Approach 2:
The patent performs energy transformation and contrast phase classification as preliminary post-processing steps after a single scan. By preparing multiple energy level images and tissue characterization data in advance from the single-energy acquisition, the system eliminates the need for additional scanning time while maintaining comprehensive tissue characterization capabilities
Data Source
AI summary
Various methods and systems are provided for transforming images from one energy level to another. In an example, a method includes obtaining an image at a first energy level acquired with a single-energy computed tomography (CT) imaging system, identifying a contrast phase of the image, entering the image as input into an energy transformation model trained to output a transformed image at a second energy level, different than the first energy level, the energy transformation model selected from among a plurality of energy transformation models based on the contrast phase, and displaying a final transformed image and/or saving the final transformed image in memory, wherein the final transformed image is the transformed image or is generated based on the transformed image.