AI Bladder Ultrasound Measurement With Automatic Caliper Placement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasound systems face challenges in accurately measuring bladder volume on touchscreen devices due to precision issues with caliper placement, especially on devices with varying screen sizes and limited processing capabilities, leading to inaccuracies and complexity in bladder volume determination.
Innovation Solution
The implementation of an AI model to automatically identify and predict the view of a bladder on ultrasound images, allowing for precise placement of calipers and calculation of bladder volume, even on devices with limited processing power, by employing a trained AI model to classify and segment bladder boundaries and apply caliper sets accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual caliper placement is used on touchscreen devices, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The system automatically identifies the bladder boundary and places calipers without requiring manual user input. The AI model processes the ultrasound image, detects the bladder contour, and autonomously positions measurement markers, eliminating the need for operators to manually place calipers on touchscreen devices.
Solution Approach 2:
The patent replaces manual mechanical touchscreen interaction with an automated image processing system. Instead of using fingers to place calipers on a touchscreen, the system uses computer vision algorithms to automatically detect anatomical structures and place measurement markers based on detected boundaries.
2Measurement precision
If automated caliper placement is implemented, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces an intermediate AI processing layer between the ultrasound image acquisition and the measurement display. This intermediary system handles the complex tasks of boundary detection and caliper placement, isolating the complexity from the user interface and keeping the overall system architecture manageable through modular design.
3Productivity
If AI model processing is used, then productivity is improved, but use of energy increases
Solution Approach 1:
The AI model is trained in advance on a dataset of ultrasound images and corresponding ground truth annotations. This preliminary training phase allows the model to learn bladder boundary characteristics beforehand, enabling fast inference during actual measurements without requiring excessive computational resources in real-time operation.
Data Source
AI summary
A method for automatically measuring a bladder on an ultrasound image feed comprises deploying an AI model which is trained so that when the AI model is deployed, the computing device identifies and predicts a view of the bladder, from one of a sagittal view and a transverse view; and when a new ultrasound image is processed, using the AI model identifies and predicts the view of the bladder, wherein the prediction is an AI model output; wherein if the AI model output predicts a sagittal view of the bladder, automatically applying one caliper set along a superior-inferior dimension of the bladder and measuring the superior-inferior dimension; and if the AI model output predicts a transverse view of the bladder, applying two caliper sets along each of a length and a width of the bladder and measuring the length and width thereof.