Catheter Image Recognition Using Two-Stage AI Region Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In intravascular ultrasound methods, the alternation of treatment and diagnosis devices due to blood vessel diameter restrictions increases surgical time, burdening patients and medical staff, necessitating immediate information provision and efficient image analysis with limited calculation resources.
Innovation Solution
A computer program and information processing method utilizing learning models to estimate and recognize objects in medical images obtained by a catheter, enabling immediate recognition of objects related to diagnosis assistance by analyzing medical images in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If treatment device and diagnosis catheter are alternately inserted due to blood vessel diameter restrictions, then device operation can be performed, but surgical time increases
Solution Approach 1:
The system performs preliminary image analysis and object recognition during the diagnosis catheter insertion phase, before the treatment device is inserted. By pre-processing the medical images and identifying relevant objects in advance, the system enables faster transition to treatment operations without waiting for complete sequential analysis, thus reducing overall surgical time while maintaining operational capability.
2Device complexity
If medical image analysis is performed with limited calculation resources, then portability is improved, but analysis speed decreases
Solution Approach 1:
The image analysis process is segmented into multiple stages: first, a detection model identifies potential objects of interest in the medical image; second, a recognition model analyzes only the extracted regions containing these objects. This segmentation approach reduces the computational burden by processing only relevant portions of the image rather than the entire image, enabling fast analysis with limited calculation resources while maintaining high analysis speed.
Solution Approach 2:
The system extracts and focuses analysis only on the specific regions containing objects of interest from the medical image. By taking out and analyzing only the necessary portions rather than processing the complete image data, the system achieves fast analysis speed with limited computational resources, making the system portable and efficient for use during intravascular procedures.
3Loss of time
If immediate information provision is achieved through real-time image analysis, then diagnosis assistance is improved, but calculation resource consumption increases
Solution Approach 1:
The system applies different processing qualities to different parts of the medical image. Instead of uniformly processing the entire image, it concentrates computational resources on specific local regions where objects of interest are detected. This local quality approach enables immediate information provision for diagnosis assistance while minimizing overall calculation resource consumption by avoiding unnecessary processing of irrelevant image areas.
Data Source
AI summary
A non-transitory computer-readable medium (CRM) storing computer program code executed by a computer processor that executes a process of acquiring a medical image generated based on a signal detected by a catheter inserted to a lumen organ, estimating a position of an object at least included in the acquired medical image by inputting the medical image to a first learning model for estimating a position of an object included in the medical image, extracting from the medical image an image portion by using the estimated position of the object as a reference, and recognizing the object included in the extracted image portion by inputting the image portion to a second learning model for recognizing an object included in the image portion.