Neural Network Metal Artifact Detection in CT Scout Scans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical images acquired by CT scanners often suffer from metal artifacts caused by implants, leading to degraded image quality and requiring manual inspection, which is time-consuming and prone to human error, and traditional 2D scout scans are inadequate for tube current modulation and soft tissue organ delineation.
Innovation Solution
An information processing method using a trained neural network to detect metal artifacts in lower-radiation dose three-dimensional image data from a scout scan, enabling automatic artifact reduction and improved image quality, and incorporating a 3D scout scan for enhanced tube current modulation and radiation dose optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection is used to detect metal artifacts, then diagnostic accuracy can be maintained, but processing time increases and human error occurs
Solution Approach 1:
The patent replaces the manual mechanical inspection process with an automated neural network-based detection system. The neural network automatically identifies metal artifacts in CT images, eliminating the need for radiologists to manually inspect each image while maintaining high detection accuracy. This substitution of automated intelligence for manual labor resolves the contradiction between diagnostic accuracy and processing time.
2Object-affected harmful factors
If projection-based MAR algorithms are applied retrospectively, then metal artifacts can be reduced, but redundant images are reconstructed wasting computational power
Solution Approach 1:
The patent applies preliminary action by detecting metal artifacts in the scout view before the main CT scan is performed. The neural network analyzes the low-dose scout images to identify the presence and location of metal implants, allowing the system to pre-configure artifact reduction parameters and avoid reconstructing unnecessary images. This preliminary detection prevents wasted computational resources on scans that will require extensive post-processing.
3Ease of operation
If traditional 2D scout scan is used, then patient positioning can be assisted, but tube current modulation and soft tissue organ delineation are sub-optimal
Solution Approach 1:
The patent transitions from traditional 2D scout scanning to 3D volumetric imaging for the scout view. This dimensional enhancement provides comprehensive spatial information about the patient's anatomy, enabling accurate tube current modulation in three dimensions and improved delineation of soft tissue organs. The 3D data structure maintains ease of operation for positioning while dramatically improving the precision of subsequent imaging parameters.
Data Source
AI summary
An information processing method processes an x-ray image including the steps of: obtaining first lower-radiation dose three-dimensional image data during a first scan of a patient; and detecting, using a trained neural network, a presence of an artifact (e.g., a metal artifact) in the first lower-radiation dose three-dimensional image data. An information processing apparatus includes processing circuitry for performing the detection method, and computer instructions stored in a non-transitory computer readable storage medium cause a computer processor to performing the detection method.