Automated Tube Current Modulation for Clinical Task Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated tube current modulation (ATCM) in computed tomography (CT) imaging relies solely on image noise, failing to effectively model system resolution, noise texture, and task-related information, leading to suboptimal image quality that is subjective and not adaptable to different clinical tasks or radiologists' preferences.
Innovation Solution
A method that calculates a desired dose level and generates a scan protocol based on clinical tasks, patient size, and imaging system characteristics, using machine learning to optimize tube current modulation and update an optimization model based on user feedback for personalized image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If automated tube current modulation (ATCM) relies solely on image noise to set dose operating point, then the system is simple and automated, but image quality is suboptimal and not adaptable to different clinical tasks or radiologists' preferences
Solution Approach 1:
The system dynamically adapts the ATCM strategy based on the selected clinical task. Different clinical tasks (e.g., lung cancer screening, abdominal imaging) have different image quality requirements, and the system adjusts the modulation parameters accordingly. This allows the automated system to be flexible and adaptable to various diagnostic needs while maintaining automation.
Solution Approach 2:
The system changes key parameters such as the image quality index weighting and dose constraints based on the clinical task selection. By modifying these parameters dynamically, the system can optimize image quality for different diagnostic scenarios without sacrificing automation, thereby resolving the contradiction between automation and adaptability.
2Device complexity
If ATCM uses only pixel noise as constraint, then the system is simple to implement, but it fails to model system resolution, noise texture, and task-related information
Solution Approach 1:
The system uses a composite image quality index that combines multiple factors: pixel noise, system resolution (MTF), noise texture (NPS), and task-related information. This composite approach integrates multiple quality attributes into a unified metric, enabling comprehensive image quality assessment while maintaining manageable system complexity through structured integration.
Solution Approach 2:
The enhanced image quality index serves multiple functions simultaneously: it accounts for basic noise levels, system resolution characteristics, noise texture patterns, and task-specific requirements. This multi-functional index improves measurement precision without proportionally increasing device complexity, as it provides comprehensive quality assessment through a unified framework.
3Ease of operation
If the targeted image quality level is based on radiologists' subjective preference of image noise level, then the system is easy to operate, but it does not offer flexibility for optimized performance for different clinical tasks
Solution Approach 1:
The system performs preliminary configuration by pre-defining clinical task profiles that encapsulate optimal image quality requirements for different diagnostic scenarios. When a radiologist selects a clinical task, the corresponding pre-configured parameters are automatically applied, providing both ease of operation (simple task selection) and adaptability (optimized settings for each task) without requiring manual adjustment of complex parameters.
Data Source
AI summary
Methods and systems are provided for automated tube current modulation (ATCM). In one embodiment, a method for an imaging system comprises: calculating a desired dose level based on a clinical task, a size of a subject to be scanned, and an image quality level; generating a scan protocol based on a relation between the image quality level and at least one characteristic of the imaging system; and performing a scan of the subject at the desired dose level and according to the generated scan protocol. In this way, ATCM can be optimized based on elements that impact image quality such as the clinical task, the physical characteristics of the imaging system, the individual patient, and the reader's preference.