Generative AI Needle Path Planning on CT Organ Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Preoperative needle insertion path planning in surgeries is challenging due to the reliance on two-dimensional CT images, which makes it difficult to identify vital organs and blood vessels, is time-consuming, and highly dependent on doctor expertise, leading to potential complications and subjective opinions.
Innovation Solution
An assisted surgical planning system utilizing a generative AI module to generate and adjust needle insertion paths based on organ segmentation images, allowing doctors to interactively select and confirm paths, reducing subjective influence and improving path accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If doctors perform needle insertion path planning based on two-dimensional CT images, then the planning process is familiar and easy to operate, but it is time-consuming and highly dependent on doctor expertise
Solution Approach 1:
The patent introduces an intermediary system consisting of a computing device with machine learning models that processes CT images and generates needle insertion path suggestions. This intermediary acts as a bridge between the raw medical images and the doctor's decision-making, automatically performing the time-consuming analysis while providing results that assist rather than replace the doctor's expertise.
Solution Approach 2:
The system performs preliminary analysis of CT images to identify vital organs, hard-to-identify nerves, and small blood vessels before the doctor begins path planning. By pre-processing the images and generating initial path suggestions, the system reduces the time the doctor needs to spend on routine analysis while maintaining ease of operation through familiar workflows.
2Productivity
If doctors rely on their own experience and expertise for path planning, then the planning process is quick and straightforward, but the results are subject to subjective opinions and potential errors
Solution Approach 1:
The system provides feedback to doctors by generating multiple suggested needle insertion paths based on CT image analysis, allowing doctors to compare different options and make more informed decisions. The feedback mechanism includes visualizing identified vital organs and nerves, and providing risk assessments that objective supplement the doctor's subjective judgment.
Solution Approach 2:
The system creates an accurate digital copy of the patient's anatomy from CT images, including three-dimensional reconstructions and segmented views of vital organs and structures. This digital copy serves as a reference that reduces subjectivity by providing consistent, reproducible anatomical representations that can be analyzed multiple times without variation.
3Measurement precision
If doctors adjust needle insertion paths and angles before surgery, then the path accuracy can be improved, but the process becomes time-consuming and requires multiple CT scans
Solution Approach 1:
The system performs comprehensive path analysis and generates multiple suggested insertion paths before the surgery begins, allowing doctors to make adjustments and selections in advance. By conducting the analysis preliminarily rather than iteratively during surgery, the system achieves high path accuracy without requiring multiple CT scans and repeated adjustments.
4Difficulty of detecting and measuring
If doctors use contrast-enhanced computed tomography (CECT) method to identify blood vessels, then the visibility of blood vessels is improved, but the complexity of the procedure increases
Solution Approach 1:
The patent introduces an image processing intermediary that uses machine learning algorithms to automatically identify and highlight blood vessels, vital organs, and nerves in standard CT images. This intermediary eliminates the need for contrast-enhanced procedures by computationally enhancing the visibility of critical structures through pattern recognition and image analysis.
Data Source
AI summary
An assisted surgical planning system includes a host computer, a discriminative AI module, a user navigation interface, and a generative AI module. The discriminative AI module is electrically connected to the host computer, and is configured to form plural organ segmentation images based on plural CT images. The user navigation interface is electrically connected to the host computer, and is configured to display the organ segmentation images, an initial needle insertion path, plural suggested needle insertion paths, and a final needle insertion path. The generative AI module is electrically connected to the host computer, such that the host computer connects the user navigation interface to the generative AI module. The generative AI module is configured to generate the suggested needle insertion paths based on the organ segmentation images and the initial needle insertion path, and display the suggested needle insertion paths in the user navigation interface.