Liver Resection Planning Using Hepatic Zone Complexity Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing liver resection (LR) complexity assessment relies on qualitative methods or user interaction, lacking an objective and automated approach, which hinders standardization and safety in liver surgery planning.
Innovation Solution
A computer-implemented method for liver resection planning that includes image processing using pre-trained neural networks for segmentation, pruning venous vessels to define a primary hepatic zone, and determining quantitative features to predict LR complexity using a classification model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If qualitative readings or user interaction methods are used for LR complexity assessment, then surgeon expertise can be utilized to anticipate difficulty, but the method lacks objectivity and cannot be standardized or reproduced
Solution Approach 1:
The patent replaces the mechanical/manual system of qualitative surgeon assessment with an automated computer-implemented method using machine learning models and image processing algorithms to objectively evaluate LR complexity from medical images, eliminating subjectivity while maintaining high accuracy
Solution Approach 2:
The system enables self-service assessment by automatically processing medical images and generating complexity scores without requiring surgeon interaction, allowing the algorithm to independently perform the assessment task that previously required expert human judgment
2Reliability
If automated methods are introduced to improve standardization, then objectivity and reproducibility increase, but the complexity of the assessment system increases
Solution Approach 1:
The patent segments the complex assessment task into distinct modular components: image preprocessing, feature extraction, machine learning classification, and complexity scoring. This modular architecture improves reproducibility while managing system complexity through organized, independent modules that can be developed and validated separately
3Manufacturing precision
If detailed venous vessels segmentation is performed to define primary hepatic zone, then surgical planning precision improves, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary segmentation of venous vessels and defines the primary hepatic zone before the actual complexity assessment. This pre-processing step creates reusable anatomical references that speed up subsequent assessments while maintaining high precision in surgical planning
Solution Approach 2:
The system performs segmentation at an appropriate level of detail - sufficient to define the primary hepatic zone accurately for surgical planning, but not excessively detailed to the point of unnecessary computational burden. The segmentation is tailored to the specific requirements of LR complexity assessment
Data Source
AI summary
The invention relates to a computer-implemented method of liver resection surgery planning, comprising the steps of: processing (E2) preoperative computed tomographic images (10) of a patient to generate liver, liver lesion (ss) and venous vessels segmentations; pruning the venous vessels segmentation to retain only major vessels and determining (E3) a primary hepatic zone (PHZ) from the retained major vessels; determining (E4) at least one quantitative feature from the primary hepatic zone, the liver segmentation and the liver lesion(s) segmentation; processing (E5) the determined quantitative features with a classification model to determine a liver resection complexity score for the patient.