Ultrasound Imaging System Automated Protocol Compilation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ultrasound fetal screening technologies face challenges in standardizing 2D and 3D image acquisition, particularly in the first trimester, due to fetal position and motion, leading to suboptimal image selection and lack of context for retrospective review, which limits quality control and training.
Innovation Solution
A computer-implemented system that receives a time sequence of 2D and/or 3D ultrasound images, registers them with a model of the anatomical body, identifies the best images according to a predetermined protocol, and compiles a standardized video sequence, including video snippets with contextual information, to facilitate standardized scanning and remote review.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated image selection and protocol compilation are implemented, then quality control and standardization are improved, but device complexity increases
Solution Approach 1:
The system performs automated image selection, protocol compilation, and report generation without requiring manual intervention from sonographers. The computer automatically registers ultrasound images with anatomical models, selects optimal images according to predetermined protocols, and compiles standardized video sequences, enabling the system to serve itself and reduce reliance on skilled operators for quality control tasks.
Solution Approach 2:
The patent replaces manual mechanical processes (physical image selection, manual protocol following, handwritten reports) with automated computational processes. The computer system uses image registration algorithms, automated image selection based on protocol criteria, and electronic report generation to substitute the mechanical and manual operations previously performed by sonographers.
2Loss of information
If comprehensive contextual information is recorded for all images, then information completeness is improved, but data volume and processing requirements increase
Solution Approach 1:
The system extracts only the most relevant contextual information from the ultrasound examination process. Instead of recording all possible data, the computer selectively captures protocol-specific images, key anatomical landmarks identified through image registration, and essential measurement data. This extraction approach preserves critical contextual information while minimizing unnecessary data volume.
Solution Approach 2:
The system performs preliminary image registration with anatomical models before final image selection and compilation. By pre-registering images and identifying key anatomical structures in advance, the system prepares contextual information that will be needed for protocol compliance and report generation, reducing the need for extensive post-processing and minimizing redundant data storage.
3Productivity
If manual image selection by skilled sonographers is reduced, then productivity is improved, but measurement precision may worsen
Solution Approach 1:
The system incorporates feedback mechanisms where the automated image selection process is continuously refined based on protocol requirements and anatomical model registration results. The computer evaluates selected images against predetermined protocol criteria and anatomical accuracy standards, providing feedback that guides further image selection and ensures measurement precision is maintained even as manual intervention is reduced.
Solution Approach 2:
The system creates accurate digital copies and representations of the ultrasound images through registration with standardized anatomical models. These digital copies preserve the essential diagnostic information and anatomical relationships, allowing automated analysis and measurement without requiring continuous manual verification, thus maintaining precision while improving productivity.
Data Source
AI summary
A system for generating ultrasound data in respect of an anatomical body being scanned makes use of a predetermined scan protocol, which specifies a sequence of types of ultrasound scan of structures of interest. These types may for example be different imaging modalities (static, dynamic, 2D, 3D etc.) which are most appropriate for viewing different structures within the anatomical body. From received ultrasound images, a model is used to identify the structures of interest within the ultrasound images. The best images for creating the types of scan of the protocol are then identified and a sequence is compiled of those best images. In this way, a sequence is created which combines different types of scan, in a structured way according to a predetermined protocol. This makes the analysis of the sequence most intuitive for a user, and simplifies comparison between different sequences.


