Medical Image Dataset Splicing for Diagnostic Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical imaging devices generate large amounts of data, leading to storage and transmission challenges due to high compression rates that compromise image quality, necessitating a method to reduce data while preserving diagnostic quality.
Innovation Solution
A system and method for splicing medical image datasets by segmenting and combining images using masks to identify points of interest and nearby areas, allowing for the creation of a single spliced image that retains optimal image quality from multiple datasets with different reconstruction parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple medical image datasets with different reconstruction parameters are stored, then diagnostic quality is improved, but storage space and transmission time increase
Solution Approach 1:
The patent combines multiple medical image datasets with different reconstruction parameters into a single spliced image dataset. The system identifies complementary regions across datasets (e.g., lung structures from one kernel, soft tissue from another) and merges them into one unified dataset that provides diagnostic quality equivalent to multiple separate datasets while reducing storage requirements.
Solution Approach 2:
The patent applies different reconstruction parameters to different regions of the image based on diagnostic needs. For example, a hard kernel is applied to lung regions where sharp edges are needed to distinguish air from non-air structures, while a soft kernel is applied to soft tissue regions where noise reduction is prioritized. This local optimization maintains diagnostic quality without requiring full datasets for all regions.
2Quantity of substance
If data compression schemes are applied, then storage space is reduced, but image quality deteriorates
Solution Approach 1:
The patent performs preliminary processing by creating a spliced image dataset that pre-identifies and preserves diagnostically important regions before storage or transmission. By segmenting and splicing datasets in advance to retain only relevant information, the system eliminates the need for aggressive compression that would compromise quality, as the spliced dataset already contains optimized diagnostic information.
3Reliability
If multiple reconstruction kernels are used, then diagnostic information is enhanced, but transmission time increases
Solution Approach 1:
The patent merges multiple reconstruction datasets into a single spliced image that consolidates diagnostic information from different kernels. This unified dataset can be transmitted as one file rather than multiple separate files, significantly reducing transmission time while preserving all essential diagnostic information needed for clinical decision-making.
Data Source
AI summary
A system and method for splicing medical image datasets are provided. The method for splicing medical image datasets comprises: segmenting first and second medical image datasets comprising an organ of interest and a nearby area to create a mask for points in the first and second medical image datasets, wherein the mask identifies points in the organ of interest and nearby area; and creating a spliced image of the first and second medical image datasets by using the points in the organ of interest and nearby area identified by the mask.