AI Landmark Localization for MRI Scan Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI scan planning is user-dependent and time-consuming, relying on experienced operators to identify anatomical landmarks from sparse scout images, which slows down the acquisition of diagnostic-quality images.
Innovation Solution
A method utilizing a machine learning model to dynamically adapt the acquisition process of survey scans by predicting anatomical landmarks, allowing for automatic planning of high-resolution MRI scans without operator intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If user intervention is used to identify anatomical landmarks from sparse scout images, then the accuracy of landmark identification can be maintained, but the scan planning time increases significantly
Solution Approach 1:
The system enables self-service by implementing an automated machine learning model that independently identifies anatomical landmarks without requiring operator intervention. The model processes survey scan images and automatically detects landmarks, eliminating the need for manual identification while maintaining diagnostic accuracy.
Solution Approach 2:
The patent replaces the manual mechanical process of landmark identification with an automated computational system. The machine learning model substitutes the operator's visual inspection and manual marking process, using algorithmic image analysis to detect landmarks automatically from survey scan data.
2Reliability
If experienced operators perform scan planning manually, then the quality of scan planning can be ensured, but the productivity of the imaging system decreases
Solution Approach 1:
The system performs scan planning autonomously without requiring experienced operators. The machine learning model independently executes the complete scan planning workflow, including survey scan acquisition, landmark identification, and high-resolution scan parameter determination, thereby freeing operators for other tasks and increasing overall system productivity.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between the survey scan data and the final scan planning decisions. This intermediary automatically processes the intermediate steps of landmark identification and parameter optimization, bridging the gap between raw survey data and diagnostic scan protocols without human intervention.
3Extent of automation
If rule-based acquisition is used for survey scans, then the process can be automated, but the adaptability to different anatomical variations is limited
Solution Approach 1:
The system dynamically adjusts acquisition parameters based on the specific anatomical features detected in each survey scan. The machine learning model analyzes the actual anatomical variations present and automatically modifies scan parameters such as field of view, slice thickness, and imaging planes to optimize visualization of the relevant anatomy, thereby achieving both automation and adaptability.
Solution Approach 2:
The patent implements a dynamic scan planning approach where the acquisition parameters are not fixed but adapt based on the detected anatomical landmarks. The system continuously adjusts imaging parameters according to the specific anatomical configuration identified by the machine learning model, enabling flexible adaptation to various anatomical presentations while maintaining automated operation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure relates to a method for medical imaging method for locating anatomical landmarks of a predefined anatomy. The method comprises: a) providing a machine learning model for predicting anatomical landmarks in image data obtained using a set of acquisition parameters and for predicting a subsequent set of acquisition parameters of the set of acquisition parameters for subsequent acquisition of image data; b) determining 5 a current set of acquisition parameters;c) receiving survey image data representing a slice of the anatomy, the survey image data having the current set of current acquisition parameters; d) identifying anatomical landmarks in the received image data using the machine learning model; e) predicting another set of acquisition parameters using the machine learning model and repeating steps c)-e) for a predefined number of repetitions using the predicted set of 10 acquisition parameters as the current set of parameters; and f) providing the identified anatomical landmarks.