Automated Lung Functional Contour Delineation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for delineating lung functional volumes on functional images are unreliable, time-consuming, and lack reproducibility, making them unsuitable for clinical use in radiation therapy planning due to variability in threshold values and sensitivity to noise and hot spots in images.
Innovation Solution
A fully automated method that aligns anatomical and functional images, determines lung anatomical contours, and iteratively varies thresholds to calculate functional contours based on total radioactivity ratios, providing consistent and clinically meaningful results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual delineation of functional volumes is performed on functional images, then the functional contours can be obtained, but the process is time-consuming and has low repeatability and reproducibility
Solution Approach 1:
The system performs automatic delineation of functional lung volumes using algorithm-based processing of functional images and anatomical images, eliminating the need for manual contouring by operators. The computer automatically calculates functional contours based on predetermined thresholds and image data, making the system self-sufficient for the delineation task.
Solution Approach 2:
The manual mechanical process of drawing contours by operators is replaced with an automated computational system that processes functional images and anatomical images through algorithms. The system substitutes human manual operations with computer-based image processing and automatic contour calculation.
2Extent of automation
If fixed threshold segmentation is used to delineate lung functional volumes, then the process becomes automated, but the results are inconsistent due to wide variety of thresholds and sensitivity to noise and hot spots
Solution Approach 1:
The system uses feedback from both functional images and anatomical images to determine the delineation threshold. The anatomical image provides structural context that feeds back into the threshold selection process, allowing the system to adjust the threshold based on the actual anatomical boundaries observed in the images, thereby improving consistency.
Solution Approach 2:
The system dynamically adjusts the threshold parameter based on the specific characteristics of each patient's images. Rather than using a fixed predetermined threshold, the system modifies the threshold value according to the intensity distribution in functional images and the corresponding anatomical structures, ensuring consistent and reliable delineation across different cases.
3Ease of manufacture
If fixed percentage threshold of maximum intensity is used, then the delineation is simplified, but the results are unreliable due to hot spots and noise in functional images
Solution Approach 1:
The anatomical image serves as an intermediary that mediates between the functional image data and the threshold determination. By comparing functional image intensities with corresponding anatomical structures, the system identifies reliable threshold values that are not influenced by hot spots or noise in the functional images alone.
Solution Approach 2:
The system performs preliminary analysis of both functional and anatomical images to establish the threshold before final delineation. This preliminary step involves examining the intensity distribution and anatomical boundaries to determine an appropriate threshold that avoids the influence of hot spots and noise, ensuring accurate subsequent delineation.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Method and device for delineating lung functional contours on functional images comprising : (a) aligning an anatomical image and a functional image such that both images have corresponding coordinates; (b) determining a lung anatomical contour on the anatomical image; (c) transferring the lung anatomical contour on the functional image; (d) obtaining the total radioactivity inside the lung anatomical contour "A0"; (e) determining a functional contour on the functional image with a threshold k such that all pixels inside said functional contour have a pixel value higher than the threshold; (f) obtaining the total radioactivity inside the functional contour "Ak"; (g) obtaining a functional value corresponding to the ratio between the total radioactivity inside the functional contour and the total radioactivity inside the anatomical contour; (h) storing the function contour if its functional value has a functional value (FV) expected.