Automatic ROI Generation on 3D Patient Surfaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing patient motion tracking systems in radiotherapy require manual drawing of the region of interest (ROI), which is time-consuming and prone to errors, necessitating clinician training and prolonging treatment time.
Innovation Solution
A system for automatic generation of ROI on a 3D patient surface using 3D scanning reconstruction and machine learning, incorporating stored ROI descriptive data and a quality module to ensure accurate and efficient tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual drawing of ROI is used by clinicians, then the ROI can be customized for each patient, but the process is time-consuming and requires clinician training
Solution Approach 1:
The system performs preliminary actions by automatically generating ROI proposals based on pre-stored descriptive data and machine learning algorithms before the clinician needs to review or adjust them. This preliminary automated generation eliminates the time-consuming manual drawing process while maintaining accuracy through subsequent clinician verification.
Solution Approach 2:
The system enables self-service by allowing the ROI generation process to serve itself through automated algorithms. The machine learning model automatically identifies and segments the region of interest without requiring manual intervention, though clinician approval is maintained for quality assurance.
2Reliability
If manual drawing of ROI is used, then clinicians can adjust the ROI based on their knowledge, but the process requires training and practice
Solution Approach 1:
The system introduces an intermediary automated ROI generation process between the patient data and the final ROI selection. This intermediary uses machine learning to propose ROIs that clinicians can then review and adjust, reducing the complexity of manual drawing while maintaining reliability through the clinician's final approval.
Solution Approach 2:
The system replaces the mechanical process of manual ROI drawing with an automated computational system. Machine learning algorithms substitute for the manual mechanical action of drawing, reducing the need for clinician training while maintaining ROI quality through automated generation and clinician review.
3Manufacturing precision
If manual ROI drawing is performed, then the ROI can be optimized for each anatomical site, but it prolongs treatment time
Solution Approach 1:
The system changes parameters by using machine learning models trained on specific anatomical sites to automatically generate site-optimized ROIs. The system adjusts ROI parameters automatically based on the patient's anatomy, maintaining manufacturing precision while significantly reducing the time required compared to manual drawing.
Solution Approach 2:
The system performs preliminary automated ROI generation for each anatomical site using pre-trained machine learning models. This preliminary action provides site-optimized ROIs immediately, allowing clinicians to review and approve quickly, thereby maintaining precision while improving treatment throughput.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure relates to a patient motion tracking system for automatic generation of a region of interest on a 3D surface of a patient positioned in a radiotherapy treatment room. More particularly, the disclosure relates to an assistive approach of a motion tracking system, by which a region of interest (ROI) is automatically generated on a generated 3D surface of the patient. Furthermore, a method for automatically generating a ROI on the 3D surface of the patient is described. In particular, all the embodiments refer to systems integrating methods for automatic ROI generation in a radiotherapy treatment setup.