CT Table Positioning via Deep Learning Organ Center
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In CT scanning, manually operating the patient table to position the organ of interest at the geometric rotation center of the medical imaging device results in low operation efficiency and movement accuracy.
Innovation Solution
A medical imaging method and apparatus that utilize deep learning to acquire a region of interest on a scout image, determine the geometric center of the organ of interest, and automatically move the patient table to align this center with the geometric rotation center of the medical imaging device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual operation of the patient table is used to position the organ of interest, then the operator can control the movement, but the operation efficiency and movement accuracy are low
Solution Approach 1:
The patent replaces the manual mechanical operation system with an automated computer-controlled system. The control unit receives coordinates of the geometric center of the organ of interest and automatically calculates and executes the patient table movement, substituting human manual control with an automated mechanical control system that achieves both high precision and efficiency
Solution Approach 2:
The system enables the patient table to automatically position itself based on pre-calculated coordinates. The control unit autonomously determines the movement distance and direction, and the patient table self-adjusts to bring the organ of interest to the geometric rotation center without requiring continuous manual intervention
2Manufacturing precision
If the organ of interest is not positioned at the geometric rotation center, then the scanning can be performed, but the image quality and resolution are reduced
Solution Approach 1:
The patent performs preliminary positioning calculations before the actual scanning process. The system pre-determines the geometric center coordinates of the organ of interest and calculates the required patient table movement in advance, allowing the organ to be positioned at the geometric rotation center before scanning begins, thereby ensuring optimal image quality without adding complexity to the scanning itself
Solution Approach 2:
The patent introduces a control unit as an intermediary between the positioning system and the patient table. This intermediary component receives coordinate information, calculates the optimal movement parameters, and translates them into control signals for the patient table, simplifying the overall system architecture while achieving precise positioning
Data Source
AI summary
Embodiments of the present application provide a medical imaging method and apparatus. The method includes acquiring a region of interest on a scout image of a subject using a deep learning method, determining a geometric center of an organ of interest of the subject according to the region of interest, and, according to the geometric center of the organ of interest and a geometric rotation center of a medical imaging device, driving a table to move. According to the embodiments of the present application, automatic movement of the table is achieved so that the geometric center of the organ of interest is close to the geometric rotation center of the medical imaging device, enabling diagnostic images to have higher image quality, including a higher image resolution. In addition, rays passing through the geometric rotation center of the medical imaging device are fully utilized, improving the dose efficiency of the rays.