Automated Lymph Node Image Processing for TNM Staging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The assessment of lymph nodes from PET or CT images is challenging due to complex labelling schemes, region-dependent risk boundaries, the large number of lymph nodes, and unclear nodal boundaries, leading to time-consuming and potentially inaccurate staging results.
Innovation Solution
An automatic method for processing images of subjects with lymph nodes, involving segmentation, delineation, classification, and risk assessment per nodal region, rather than per node, to facilitate fast and accurate TNM staging and reduce reader dependencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual detection and assessment of lymph nodes is performed, then accuracy of lymph node identification can be improved, but time consumption increases significantly
Solution Approach 1:
The patent segments the complex task of lymph node assessment into distinct automated steps: detection, segmentation, characterization, and staging. This segmentation allows computer-aided assessment to handle routine tasks automatically while radiologists focus on complex cases, resolving the contradiction between accuracy and time consumption.
Solution Approach 2:
The patent introduces a computer-aided assessment system as an intermediary between the images and the final staging report. This intermediary performs automated analysis to generate preliminary results, which radiologists then review and refine, thereby maintaining high accuracy while significantly reducing overall assessment time.
2Measurement precision
If comprehensive assessment of all lymph nodes is performed, then staging accuracy is improved, but workload and time pressure increase
Solution Approach 1:
The patent applies local quality by performing detailed automated assessment only where needed - the system characterizes detected lymph nodes and applies region-specific risk boundaries automatically. This allows comprehensive staging accuracy without manually evaluating every single lymph node, as the automated system handles the detailed local analysis.
Solution Approach 2:
The patent changes the assessment parameters from individual lymph node evaluation to region-based risk assessment. By aggregating lymph node characteristics into regional risk scores and applying standardized risk boundaries, the system achieves comprehensive staging without the linear workload increase of assessing each node individually under time pressure.
3Measurement precision
If detailed lymph node characterization is performed, then diagnostic accuracy is improved, but complexity of the assessment process increases
Solution Approach 1:
The patent performs preliminary automated characterization of all detected lymph nodes before radiologist review. The system pre-calculates size measurements, texture features, and metabolic activity metrics, so that when radiologists review cases, the detailed characterization work is already completed. This maintains high diagnostic accuracy while reducing the perceived complexity for the radiologist.
4Measurement precision
If region-dependent risk boundaries are applied, then accuracy of malignancy assessment is improved, but difficulty of detection and measurement increases
Solution Approach 1:
The patent creates a universal automated system that handles multiple region-specific risk boundary schemes simultaneously. The computer-aided assessment tool is configured with different regional risk criteria and automatically applies the appropriate scheme based on the lymph node location, making the complex region-dependent assessment as simple as selecting a preset configuration.
Data Source
AI summary
According to an aspect, there is provided a computer implemented method of processing an image of a subject comprising lymph nodes, the method comprising: segmenting lymph nodes in image data corresponding to the image, delineating lymph nodes from the segmented image data, classifying a lymph node as belonging to a predefined region, evaluating a region of the image based on an assessment of the risk of the region comprising a malign lymph node, and indicating the evaluation of the region.